Progress in Oceanography 206 (2022) 102857

Contents lists available at ScienceDirect

Progress in Oceanography
journal homepage: www.elsevier.com/locate/pocean

Long-term hydrographic changes in the Gulf of California and ecological

impacts: A crack in the World’s Aquarium?
William Gilly a, *, Unai Markaida b, Patrick Daniel a, c, Tim Frawley a, d, Carlos Robinson e,
Jaime Gómez-Gutiérrez f, Dylan Hyun g, Jacob Soliman g, Puneeta Pandey a,
Lorenzo Rosenzweig h
a
Hopkins Marine Station of Stanford University, Pacific Grove, CA 93950, USA
b
El Colegio de la Frontera Sur, CONACyT, Campeche 24500, Mexico
c
University of California Santa Cruz, Santa Cruz, CA 95064, USA
d
Institute of Marine Science, University of California Santa Cruz, Santa Cruz, CA, USA
e
Instituto de Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México, Mexico City, Mexico
f
Centro Interdisciplinario de Ciencias Marinas, Instituto Politécnico Nacional, La Paz, BCS, Mexico
g
Stanford University, Stanford, CA 94305, USA
h
Terra Habitus, A.C., Garza García, Nuevo León, Mexico

A B S T R A C T

Although the Gulf of California is widely recognized as a region of high productivity and biodiversity, recent oceanographic and ecological changes have had a
significant impact on its overall health. We review the relevant history of the economically important fishery based on large Humboldt (jumbo flying) squid (Dosidicus
gigas) (>50 cm mantle-length and 10 kg body weight) that existed in the Guaymas Basin from the 1990′ s until 2009, when a strong El Niño was accompanied by
precocious spawning of squid at extremely small size (<30 cm mantle length and 0.1 kg). This short-lived phenotype is characteristic of the equatorial part of the
species range and cannot be profitably fished. This phenomenon also occurred in conjunction with the strong 1997–1998 El Niño. Although rapid recovery of squid
size and landings occurred after the 1997–1998 event, recovery after 2009–2010 El Niño was still incomplete when another strong El Niño occurred in 2015–2016.
This last event has led to fixation of the small size-at-maturity phenotype in the Gulf of California — an effective poleward migration of the tropical phenotype. We
also document years between El Niño 2009–2010 and 2015–2016 as being characterized by significant subsurface warming, leading to decreased productivity and
restricted nighttime foraging opportunities for squid in the upper water column. These climatic trends have likely altered trophic webs throughout the Gulf of
California, impacting many other taxa, including small pelagic fishes that are ecologically vital and commercially harvested in large quantities, as well as a suite of
larger teleost fishes that are targeted by sport fishers. From these disparate data sources, collected and collated in coordination with regional resource users and
environmental monitoring programs, an overall pattern emerges of long-term decline across multiple taxa. We also present evidence of a significant abatement of
subsurface warming after 2015 and consider when – or if – the Gulf of California will return to a ‘normal’ state like that prior to 2010. Despite the potential for
shifting to a cooler oceanic regime, the small phenotype for jumbo squid has persisted until present (2022), and the overall trajectory of ecological recovery is
unclear. Integrated analyses of additional oceanographic and biological time-series, fisheries records, and other data will be necessary to identify critical processes
and trajectories occurring in the Gulf of California.

1. Introduction occurring in the planet’s oceans are present in the property, giving it
extraordinary importance for the study of marine and coastal processes”
The Gulf of California, also known as the Sea of Cortez, has often (UNESCO, 2005). <15 years later, the same parts of the Gulf of Cali­
been called the “Aquarium of the World,” an epithet generally attributed fornia were added to the List of World Heritage in Danger (UNESCO,
to explorer Jacques Cousteau. Accordingly, the Protected Areas and 2019). This action primarily reflected highly politicized issues con­
Islands of the Gulf of California were named a UNESCO World Heritage cerning the potential extinction of the vaquita (Phocoena sinus)
site in 2005, because “almost all major oceanographic processes (UNESCO, 2021), the world’s smallest cetacean, and the associated

Abbreviations: OMZ , Oxygen Minimum Zone ; OLZ , Oxygen Limited Zone ; DOx20 , Depth where dissolved oxygen = 20 micromol/kg ; DOx60 , Depth
where dissolved oxygen = 60 micromol/kg ; D22 , Depth where temperature = 22 C ; ML , Mantle length

* Corresponding author.
E-mail address: [email protected] (W. Gilly).

https://doi.org/10.1016/j.pocean.2022.102857
Received 31 December 2021; Received in revised form 23 June 2022; Accepted 8 July 2022
Available online 14 July 2022
0079-6611/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Fig. 1. A) Study areas in the Gulf of California. CTD deployments were carried out in the Guaymas Basin and Southern Gulf of California, Mexico. B) white box
defines the area previously reported on for the period 2006–2015 (Frawley et al. 2019a), and the smaller dotted yellow box off Santa Rosalia encloses the section
(black rectangle) analyzed for the present study. C) Casts in this section focused on a grid of established sites (T2, T4, T10B, and T12) located approximately on the
350 and 700 m depth isobaths offshore Santa Rosalia. Another set of CTD casts followed a roughly linear transect (dotted line) off La Reforma from 40 m depth
(PP40) to 500 m depth (PP500), with additional casts at depths of 130 m and 300 m. CTD casts analyzed for the Southern Gulf of California (GOC) region were
enclosed by the section defined by the black box in panel B. Sportfish monitoring was based on landings made from two sets of pangas operating out of San José del
Cabo. Fishing efforts were spread out along the coast on either side of this port, generally within ~ 30 km of shore. Blue dots indicate locations of CTD casts in World
Ocean Database 13 and in our unpublished collection (2000–2020).

illegal gillnet fishing for endangered totoaba fish (Totoaba macdonaldi) (Hoving et al. 2013; Frawley et al., 2019a). Subsurface thermal anom­
to satisfy a black market for swim bladders (Brownell et al., 2019; Rojas- alies were also associated with decreased levels of wind-driven up­
Bracho et al., 2019; Gulland et al., 2020). Although these activities are welling and productivity in the Gulf (Robinson et al., 2013, 2016), and
limited to the northernmost portion of the Gulf of California, major the combination of these factors is thought to have reinforced a switch to
oceanographic and ecological changes have occurred elsewhere in the a phenotype that is characteristic of the tropical portion of the jumbo
gulf during the same period (Aburto-Oropeza et al., 2015; Adame et al., squid’s biogeographic distribution (e.g., Ecuador and Costa Rica) in
2020; Favoretto et al., this issue) but have received considerably less which individuals precociously mature at unusually small size (Hoving
attention. et al., 2013; Chen et al., 2013). In turn, this small size prevents squid
Jumbo flying, or Humboldt, squid (Dosidicus gigas) supports the from being viably harvested in commercial quantities using established
world’s largest invertebrate fishery and accounted for one third of all jigging methods in Mexico (Frawley et al., 2019a).
squid landings in 2019 (FAO, 2021; Rodhouse et al., 2016; Arkhipkin During the peak of the jumbo squid fishery, about 100,000 tons were
et al., 2015a). Large jumbo squid became a prominent presence in harvested from the Gulf of California annually, and an assessment of
southern half of the Gulf of California in the 1980′ s (Gilly et al., 2020) population size based on tag-and-recapture data (Markaida et al., 2005)
and supported the third largest fishery in Mexico from 1995 to 2005 resulted in estimates of ~ 20 million and ~ 130 million jumbo squid in
(behind sardine (Sardinops sagax) and yellowfin tuna (Thunnus alba­ October 2001 and April 2002, respectively (Morales-Bojórquez et al.,
cares)), when squid landings were exceeded by those of the small pelagic 2012). Such a large number of highly predatory squid would be expected
fish, Pacific thread herring (Opisthonema libertate) (FAO, 2021). Of these to exert significant ecological impacts on lower trophic levels on which
species, only squid was primarily targeted by an artisanal sector that was the squid feed, such as myctophid fishes, krill and other micronekton
composed of ~ 1800 small-scale fishing vessels in 2009 (Ramírez- (Rosas-Luis et al., 2008; Portner et al., 2019), as well as on top predators
Rodríguez and Almendárez-Hernández, 2013). Squid landings in Mexico that consume large squid, particularly sperm whales (Physeter macro­
declined precipitously in subsequent years, falling to the 20th position cephalus) and potentially sea lions (Zalophus californianus) (Jaquet and
on the single-species fisheries list by 2015. During this period, Dosidicus Gendron, 2002; Davis et al., 2007; Ruiz-Cooley et al., 2012; Adame
landings in the southern hemisphere remained high. et al., 2020). How such direct impacts were altered by the change in
This sudden crash in the Mexican squid fishery has been related to squid phenotype is largely unknown, but once-common sperm whales
climatic anomalies driven by strong El Niño events in 2009–2010 and have essentially disappeared from the central Gulf of California
2015–2016 and a period of subsurface warming between these markers following the decline of large jumbo squid (Rosenzweig, unpublished

2
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

data from Lindblad Expeditions ecotourism operations, and D. Gendron, Bellevue, WA, USA). A total of 881 CTD casts were obtained at depths >
pers. comm., La Paz, BCS, Mexico). Numbers of sea lions have declined 40 m between July 2006 and December 2020 in the area around Santa
throughout the Gulf of California over the last two decades (Adame Rosalia, BCS, Mexico (Fig. 1B). The vast majority of CTD casts (717
et al., 2020). during 2014–2020, including both up and down casts when possible)
Jumbo squid are well adapted to tolerate both temperature and were carried out at a grid of specific sites near Santa Rosalia (Fig. 1C) in
oxygen changes, and they spend most of the daytime near the upper conjunction with monitoring efforts of the SURMAR program (Sustain­
boundary of the oxygen minimum layer (OMZ), where oxygen concen­ able Utilization and Research of the Mar de Cortés) (https://oceanfdn.
tration is < 20 µmol/kg (~10 % of that at the surface). During the night, org/projects/surmar-asimar/; accessed 04/02/2022), a collaboration
jumbo squid ascend, and forage in the upper 75 m of the water column with the Instituto Tecnológico Superior de Mulegé and Minera y Met­
(Gilly et al., 2012). Thus, properties of the upper water column are alúrgica del Boleo in Santa Rosalia, BCS, Mexico. In addition, 74 casts
relevant to nighttime foraging, an activity that may supply most of the were carried out using a variety of small, private vessels (2006–2015, W.
squid’s nutrition. This pattern of diel vertical migration was observed in Gilly), and 56 were recorded on board the RV El Puma (UNAM) during
every location where squid were tagged between 2001 and 2016, supported cruises (2007–2017, C. Robinson and J Gómez-Gutiérrez).
including the Guaymas Basin in the Gulf of California (Gilly et al., 2006), Other CTD casts were obtained in collaboration with Lindblad Expedi­
Pacific Ocean off Bahia Magdalena (Bazzino et al., 2010), central Cali­ tions National Geographic Seabird and Sea Lion vessels (n = 14,
fornia (Stewart et al., 2013), and Peru (Gilly, unpublished). This 2009–2013) and during research cruises with RV New Horizon (n = 11,
behavior appears to be a universal feature of the jumbo squid’s foraging 2007–2011), RV BIPXII (n = 5, 2008–2013), and RV Pacific Storm (n = 4,
strategy throughout its broad distribution in the eastern Pacific. 2007).
In the course of the previous squid-tagging work we repeatedly A total of 239 CTD casts were collected in the southern Gulf of Cal­
observed that squid spent very little time in waters with temperatures > ifornia region (Fig. 1B) during 2008–2018 as described above by Lind­
22 ◦ C, as estimated by the fraction of time spent in near-surface waters blad Expeditions (2009–2018), and on research cruises using the vessels
warmer than this temperature in the Gulf of California (Gilly et al., 2012; Don José (n = 22, 2008–2010, Stanford University Holistic Biology
Stewart et al., 2013). Furthermore, if squid entered this warm nighttime program), RV El Puma (n = 11, 2011–2017), RV New Horizon (n = 7,
foraging zone, they remained at temperatures > 22 ◦ C for only a few 2010–2011), and RV BIPXII (n = 2, 2013). An additional 98 casts
minutes before diving to much cooler depths, presumably to relieve (2008–2017) were obtained from World Ocean Database 2018 (WOD18)
thermal stress (Davis et al., 2007). Both behavioral responses showed a on November 12, 2020 (https://www.ncei.noaa.
highly non-linear dependence on temperature, with 22 ◦ C emerging as gov/products/world-ocean-database; accessed 06/23/2022).
an apparent threshold. All hydrographic data collected by our efforts were processed using
Resting oxygen consumption by jumbo squid (under normoxic con­ the SBE Data Processing tools (Seabird Electronics) with 1 m bin-
ditions) increases very rapidly with temperature > 20 ◦ C (Rosa and averaging and then imported into Ocean Data View 4.78 (Schlitzer,
Seibel, 2008, Rosa and Seibel, 2010; Seibel, 2013), and it is likely that 2021) for analysis and graphics generation and for exporting selected
active foraging at such high temperatures is metabolically unsustain­ data as csv-files to enable further analysis using Microscoft Excel and
able. Thermal tolerance due to metabolic limitations is thus challenged Igor Pro 8.04 (Wavemetrics, Portland, OR, USA). Daily means (n = 153)
when squid attempt to forage in the productive upper water column were computed for casts made on the same day (defined by GMT) in a
above the thermocline when temperature exceeds some barrier above 20 given year in the Santa Rosalia region. Time series were smoothed using
⁰C. If squid cannot remain in such warm water for more than several a box-car method using Igor Pro as indicated in the relevant figure
minutes, they would essentially be excluded from this portion of their legends. Fits to a sine function and linear regressions were also gener­
primary nighttime foraging zone when zooplankton and micronekton ated using Igor Pro. The sine-fit function in Igor Pro uses the equation,
migrate there during their daily vertical migration (Cade and Benoit-
F(t) = y0 + Asin(f *t + Φ)
Bird, 2015).
The present study examines temporal variation of temperature and and generates values (±SD) for each parameter, with period (days) equal
dissolved oxygen concentration in the water column in the Guaymas to 2π/f and Φ in radians. Data collected in conjunction with NSF OCE
Basin region, both seasonal and interannual, based on an extensive data support have been deposited in the BCO-DMO database (https://www.
set collected over 15 years (2006–2020). These environmental changes bco-dmo.org/search/dataset/Gilly; accessed 06/23/2022). Other data
are interpreted in relation to the history of jumbo flying squid (Dosidicus are available on request.
gigas) in the same area of study over this period. We also extend the view
to include a broader area of the Gulf of California and an assortment of
fish taxa that are targeted by commercial and sport fisheries. Most fish 2.2. Sport fishing data
species targeted by artisanal fisheries in the southern Gulf of California
decreased in importance to the commercial sector and showed an Sport fishing data were obtained from two sources in Baja California
increasing catch per unit effort between 1970 and 2000 (Sala et al., Sur, Mexico. Weekly online reports of sport fish landings (1/02/2005 –
2004), but how these factors have changed since then are unclear. Based 12/31/2019) posted by Eric Bricston of the Gordo Banks Pangas fleet
on our analysis, a general pattern of long-term decline in multiple taxa operating out of Puerto Los Cabos near San José del Cabo (Fig. 1B) were
emerges over the last decade. Although the subsurface warming trend downloaded (https://gordobanks.com/category/fish-reports/; accessed
that has been associated with the shift of jumbo squid to the small, 06/23/2022), and data were imported into Excel for sorting and anal­
tropical phenotype appears to have been subsiding since 2015, it re­ ysis. Data include the number of charter vessels reporting for each week
mains unclear when (or if) large squid will again appear in the Gulf of and the numbers of each of 66 specific fishes landed (plus jumbo squid).
California, whether other declining taxa will recover, and what overall Several fish of the same type were combined into groups for analysis:
ecological state will emerge in the face of a changing midwater climate. “groupers” (gulf grouper Mycteroperca jordani, broomtail grouper
M. xenarcha, pinto (spotted) cabrilla Epinephelus analogus, baqueta (gulf
2. Materials and Methods coney) Hyporthodus acanthistius, and miscellaneous groupers/cabrillas);
“pargos” (yellow pargo Lutjanus argentiventris, colorado pargo
2.1. Hydrographic data L. colorado, huachinango L. peru, barred pargo Hoplopagrus guentherii,
and tijareta (creole fish) Paranthias colonus); and “marlins” (blue marlin
Hydrographic measurements employed a SBE19 or SBE19plusV2 Makaira nigricans and black marlin Istiompax indica). These three groups
CTD-profiler equipped with a SBE43 oxygen sensor (Seabird Electronics, and 16 other individual “species” accounted for > 90 % of all landings

3
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Table 1
Numbers of selected types of sport fish landed between 2005 and 2019 as reported for the San Jose del Cabo area of Baja California Sur, Mexico reported by Gordo
Banks Pangas (number of trips 59,391) and by Dr. John Snow (number of trips 930). Pargos* included creole fish (Paranthias colonus); Groupers** included gulf coney
(baquetta = Hyporthodus acanthistius). Slope of fit refers to the slope (±SD) of a linear regression fit to the average catch-per-trip data from both sources over all the
years (Fig. S1, Supplementary Material). Blue and black Marlins*** were not reported by John Snow, and only data from Gordo Banks Pangas were fitted in this case.
Slopes with significance of p < 0.05 are shown in bold type. Slopes with marginal significance (0.05 < p < 0.10) are in italics.
Fish Scientific name Gordo Banks (n) John Snow (n) Slope of fit r2 p Habitat
(fish/ trip.yr¡1 ± SD)

Yellowfin tuna Thunnus albacares 111,279 405 − 0.0904 ± 0.0338 0.35466 0.019 Pelagic
Dorado Coryphaena hippurus 64,953 497 − 0.0552 ± 0.0204 0.36115 0.018 Pelagic
Pargos* Lutjanus spp. 51,157 1505 − 0.0056 ± 0.0423 0.00135 0.896 Demersal
Sierra Scomberomorus sierra 41,405 189 − 0.0423 ± 0.0071 0.73124 <0.0001 Pelagic
Bonitos Sarda spp. 30,658 299 − 0.0206 ± 0.0251 0.04918 0.427 Pelagic
Black skipjack Euthynnus lineatus 24,353 444 − 0.0480 ± 0.0196 0.31585 0.029 Pelagic
Roosterfish Nematistius pectoralis 16,581 5 − 0.0035 ± 0.0057 0.02804 0.551 Demersal
Leopard grouper Mycteroperca rosacea 10,698 50 − 0.0083 ± 0.0033 0.32844 0.026 Demersal
Amberjack Seriola rivoliana 8353 38 − 0.0031 ± 0.0013 0.30176 0.034 Demersal
Wahoo Acanthocybium solandri 8247 29 0.0049 ± 0.0027 0.20281 0.092 Pelagic
Striped marlin Kajikia audax 7142 20 0.00003 ± 0.002 0.00002 0.987 Pelagic
Yellowtail Seriola lalandi 5143 50 − 0.0066 ± 0.0071 0.06260 0.368 Pelagic
Jack crevalle Caranx caninus 4878 20 − 0.0039 ± 0.0013 0.39125 0.013 Pelagic
African pompano Alectis ciliaris 3219 24 − 0.0059 ± 0.0017 0.46912 0.005 Demersal
Dog pargo Lutjanus jocu 2328 23 − 0.0071 ± 0.0024 0.40929 0.010 Demersal
Sailfish Istiophorus platypterus 2053 6 − 0.0010 ± 0.0005 0.24077 0.063 Pelagic
Rainbow runner Elagatis bipinnulata 1669 18 0.0010 ± 0.0010 0.07191 0.334 Pelagic
Groupers** Mycteroperca spp., Epinephelus spp. 1112 437 − 0.0083 ± 0.0055 0.14756 0.158 Demersal
Marlins*** Makaira nigricans, Istiompax indica 409 0 0.0008 ± 0.0002 0.52249 0.002 Pelagic
Total charter trips 59,391 930

and were used for detailed analysis: yellowfin tuna Thunnus albacares, produccion-pesquera; accessed 06/23/2022). These reports include
dorado Coryphaena hippurus, sierra Scomberomorus sierra, bonito Sarda monthly landings at specified reporting offices in all Mexican states.
chiliensis and S. orientalis, black skipjack Euthynnus lineatus, roosterfish Data for “landed weight” of each fish species were assembled for the four
Nematistius pectoralis, leopard grouper Mycteroperca rosacea, amberjack states bordering the Gulf of California (Baja California, Baja California
Seriola rivoliana, wahoo Acanthocybium solandri, striped marlin Kajikia Sur, Sonora, and Sinaloa), with landings on the Pacific coast of the Baja
audax, yellowtail Seriola lalandi, jack crevalle Caranx caninus, African California peninsula being excluded (Cabo San Lucas and Mazatlán were
pompano Alectis ciliaris, dog pargo Lutjanus jocu, sailfish Istiophorus included in Gulf of California landings). Landings for guachinango were
platypterus, and rainbow runner Elagatis bipinnulata (Table 1). The period summed with those of mixed pargos for our analysis.
of reporting included data from a total of 59,391 charter trips and Data for commercial landings of small pelagic fishes for 2005–2015
397,148 fish of the above types. A total of 433,656 fish of all 69 types (Alvarez et al., 2017) and 2006–2020 (CONAPESCA website cited
were reported during 2005–2019 as well as 252 jumbo squid records. above) were obtained from the cited sources. Average values from these
We assumed that any targeting of specific fish types by individual sources were used for years 2006–2015. Small pelagics were divided
charters was randomly distributed over the years of reporting. into two taxonomic groups for analysis – California sardine (sardina
A second set of sport fishing data was assembled by Dr. John T. Snow Monterrey in Spanish) Sardinops sagax caeruleus and five other “non-
(https://mexican-fish.com/; accessed 06/23/2022) from personal fish­ sardine” species: thread herring (sardina crinuda), Opisthonema libertate,
ing records maintained from 2003 through 2020 and entered into a round herring (sardina japonesa) Etrumeus teres, chub mackerel (mac­
logbook after each fishing trip. All fish were collected from one of two arela) Scomber japonicus), California anchovy (anchoveta norteña)
pangas in the area around San José del Cabo included in operations by Engraulis mordax, and Pacific achoveta (sardina bocona) Cetengraulis
Gordo Banks Pangas. These fishing trips were carried out with two mysticetus.
captains over the entire collecting period (2003–2008 and 2009–2020, Commercial landings of jumbo squid in the Santa Rosalia area were
respectively), and records include data for the same types of fish as obtained from the SAGARPA (Secretaria de Agricultura, GanaderÍa,
described for Gordo Banks Pangas, except for blue/black marlins. 930 Desarrollo Rural, Pesca y Alimentación) Baja California Sur office and
trips over the 2005–2019 period (same as Gordo Banks Pangas) yielded directly provided by Unai Markaida (ECOSUR). These reports contain
4,059 fish landed. Because every fish captured by John Snow was monthly values for landed weight (“peso desembarco” or “peso desem­
recorded, including many obscure species, this complete data set in­ barcado” in Spanish) and capture location of squid landed in BCS during
cludes information on a total of 52,396 individual fish of 206 species 2001–2013. Data corresponding to fishing effort (catch per unit effort)
between 2003 as well as 18 jumbo squid. Again, targeting of individual are not available. Landed weights from all reported jumbo squid land­
fish types on individual trips was assumed to be random over the years. ings in the Santa Rosalia area, regardless of landing location, were
totaled for each calendar year. These values were cross-checked with a
similar data base for 2001–2016 that was previously compiled from
2.3. Commercial fish and squid landings data monthly landings data collected by the National Fisheries and Aqua­
culture Agency (CONAPESCA) in the regional office in Santa Rosalia on
Commercial landings of several species of pelagic fishes in the Gulf of which the SAGARAPA reports are based (Frawley et al., 2019a). This
California that were also targeted by sport-fishing operations – yellowfin CONAPESA data set has recently been reissued for years 2006–2020
tuna Thunnus albacares, dorado Coryphaena hippurus, sierra Scomber­ (CONAPESCA website cited above). These CONAPESCA reports do not
omorus sierra, barrilete (black skipjack) Euthynnus lineatus, pargos include capture location, only the office at which the landings were
(mixed snapper types listed in preceding section), and guachinango registered. In general, there was excellent agreement for the Santa
(huachinango) Lutjanus peru – were obtained from data collected by the Rosalia data between these sources, except for 2012 when CONAPESCA
Mexican National Fisheries and Aquaculture Agency (CONAPESCA) data included some landings made on the Pacific coast but reported in
for years 2006–2020 (https://datos.gob.mx/busca/dataset/

4
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Santa Rosalia. There is also a major discrepancy for Santa Rosalia depth in June and shoaling to a minimum at the end of the warm period
landings in 2008 between the two CONAPESCA sources, but the total for described above. A similar variation occurs with the 60 µmol/kg O2
Baja California Sur (BCS) are in agreement. Values for Santa Rosalia and isopleth (DOx60). This pattern is out of phase with the change in tem­
BCS for 2014–2020 were taken from the newer CONAPESCA data set. perature at depth, as directly evidenced by comparing the seasonal
Values for landings for years 1995–2019 in the state of BCS were variation in temperature and dissolved oxygen concentration at a given
taken from annual reports (Anuarios in Spanish) published by CON­ depth (Fig. 1C). Dissolved oxygen concentration at 150 m depth reaches
APESCA (https://www.gob.mx/conapesca/documentos/anuario- a maximum of ~ 75 µmol/kg about 90 days prior to the temperature
estadistico-de-acuacultura-y-pesca; accessed 06/23/2022). These re­ maximum, and a similar relationship is seen at other depths (data not
ports do not contain specific capture or landing location, but most squid shown).
were caught and landed in the Santa Rosalia area until 2010 (Fig. 8B). Deployment of archival electronic tags on jumbo squid has demon­
Values for BCS during 1985–1994 were obtained from the 1995 annual strated that an individual squid spends very little time at temperatures
report, because previous publications did not include jumbo squid. It above 22 ◦ C. Sea-surface temperature reaches this level during the May-
should be noted that this report contains only “peso vivo” values and not June transition to the warm season, and the 22 ◦ C isotherm typically
landed weights as defined above. During the years 1995–2019, when reaches a depth of ~ 50 m by the end of July (Fig. 2A). This isotherm and
both “peso vivo” and landed weight were reported, landed weight was depth are of particular interest when considering jumbo squid ecology,
80.8 ± 3.3 % (mean ± SEM) of “vivo” weight for BCS landings, and we because squid typically forage in the upper 75 m of the water column at
used the reported “vivo” values with no correction in our analysis. night. Thus, we have hypothesized that if the 22 ◦ C isotherm reaches
such depths during the warm season, it creates a stressful habitat and
2.4. Squid size data may impair nighttime foraging by squid.
Seasonal variation in depth of the 22 ◦ C isotherm (D22) shows a
We sampled a total of 2,504 individual jumbo squid from the strong relationship to jumbo squid landings in Santa Rosalia during
Guaymas Basin during 2005–2020, mostly from the area around Santa 2001–2008 (Fig. 3), with historical landings reaching a peak around the
Rosalia where CTD casts were made (Fig. 1) and determined dorsal time D22 exceeds 50 m. Landings then start to decline as D22 reaches its
mantle length (ML), sex, and maturity stage (Lipiínski and Underhill, maximum during August through October. During this warm season,
1995) of each specimen. Of these squid, 1,086 were mature (i.e., Stages DOx60 shoals, reaching a minimum depth of ~ 125 m in November.
4 or 5). All squid were captured by jigging using hand-line or rod-and- Although there is no evidence that this oxygen concentration is stressful
reel, but the size of the jigs used changed during the course of the for jumbo squid, we are unaware of any data on the potential for dele­
study to facilitate capturing smaller squid. Prior to 2010, conventional terious interactions of temperature and oxygen concentration in these
weighted jigs of 30–50 cm length (like those used in the commercial ranges, e.g. ~ 15 ◦ C and 50 µmol/kg O2 at 150 m depth as recorded in
fishery) were used, but as squid became smaller after 2010 (Fig. 8A), the November (Fig. 2C).
size of conventional-style jigs was reduced to 15 cm, and we eventually
adopted 9 cm plastic jigs of a different type (Yo-Zuri Ultra Lens, Yo-Zuri 3.2. Long-term variation in hydrographic properties – Temperature
America, Port St. Lucie, FL, 34986, USA). During the years of sampling,
the fraction of mature squid in any given sample increased from ~ 0.25 Both increasing temperature and decreasing oxygen in the water
in 2005 to ~ 0.75 in 2019, either because population dynamics had column during the warm season are thus strong environmental signals
changed with the phenotype shift, or because immature squid after 2010 that were temporally related to the seasonal presence of large numbers of
were too small to be easily captured, even with the smaller jigs (Frawley jumbo squid in the Santa Rosalia area during the years before 2010 when
et al., 2019a). commercial fishing flourished. As previously reported for depths in the
Most jumbo squid (n = 921) were sampled in Santa Rosalia in 75–150 m range in the Guaymas Basin between 2006 and 2017 (Frawley
conjunction with commercial fishing operations or our ongoing research et al., 2019a), a subsurface warming trend during the July-October warm
program (2005–2019). Other squid were collected during research season also occurred in the Santa Rosalia area (Fig. 4A). Prior to 2009,
cruises on the vessels employed for CTD casts: RV El Puma (n = 760), RV which marked the beginning of a strong El Niño (2009–2010), the 20 ◦ C
New Horizon (n = 413), RV BIP XII (n = 365), and RV Pacific Storm (n = isotherm was at 50 m depth, but this feature increased to ~ 75 m depth
45). after El Niño and remained there until another strong El Niño in
2015–2016, when it was depressed even further to 100 m depth. A similar
3. Results pattern is evident for the 22 ◦ C and 25 ◦ C isotherms. After El Niño
2015–2016, all illustrated isotherms slowly recovered, but temperatures
3.1. Seasonal hydrographic changes in relation to jumbo squid landings in remained above those observed before 2010.
the Guaymas Basin These trends were analyzed more quantitatively by fitting linear
regressions to daily means of CTD data for the periods 2006–2020,
Seasonality in water-column properties in the monitored area off 2006–2015, and 2015–2020 at a series of depths ranging from 2 m to
Santa Rosalia in the central Gulf of California is characterized by a cold 300 m (Fig. 5A). Fits to the entire time series (2006–2020) did not yield
period (December-May) and a warm period (July-October) separated by significant (p < 0.05) slopes for any depth except 2 m (Table 2). Fits to
transitions in June and November when temperature changes rapidly 2006–2015 data gave positive slopes with significant values (p < 0.05)
(Fig. 2A). This pattern is evident in near-surface waters (2–5 m) as well for all depths between 25 and 200 m, and the maximum slope (0.35
as at depth, to about 350 m where temperature is about 10 ◦ C year- ˚Cyr− 1) occurred between 50 m and 75 m depth (Table 2; Fig. 5B).
round (data not shown). This cycle is similar to that observed for the Fits to the second half of the time series (2015–2020) show a sig­
much larger area of the Guaymas Basin (Fig. 1B), although in that case nificant (p < 0.05) cooling trend at all depths except 25 m, where sig­
the warming transition appears to begin a few weeks earlier in May nificance is marginal (p = 0.08; Table 2). A maximum slope of − 0.4
along the Sonora coast (data not shown). Chlorophyll a concentration in ◦
Cyr-1 occurred at 75 m (p < 0.01), and slopes for all depths between
surface waters in this region shows a roughly reciprocal seasonal 100 m and 250 m were highly significant (p ≤ 0.001) (Table 2; Fig. 5B).
pattern, increasing during the cool period to a maximum in March-April The rising and falling slopes correspond to maximum temperature
and falling to a minimum in July-September (Robinson et al., 2016). changes at 100 m depth of about + 3.0 ◦ C for 2006–2015 and − 2.0 ◦ C for
Dissolved oxygen concentration in the water column also varies 2015–2020 (Fig. 5C). The post-2015 cooling trend has thus resulted in a
seasonally (Fig. 2B), with the depth of the upper boundary of the oxygen net temperature increase in 2020 of ~ 1 ◦ C at 100 m from the level in
minimum zone (20 µmol/kg O2 = DOx20) increasing to a maximum 2006 (diamond symbols in Fig. 5C).

5
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Fig. 2. Seasonal variation in average temperature and dissolved oxygen concertation in the water column. A) Temperature profiles obtained from all CTD casts off
Santa Rosalia (see Fig. 1) during 2006–2020 are plotted versus day of year. Dashed lines indicate the transition periods in June and November (Hidalgo-González and
Alvarez-Borrego, 2004). B) Dissolved oxygen concentration at depth versus day of year during 2006–2020. The upper boundary (20 µmol/kg O2) of the oxygen
minimum zone (OMZ) (DOx20)reaches maximum depth in June and shoals to a minimum at the end of the warm season in November. The 60 µmol/kg O2 isopleth
(DOx60) varies in a similar manner. This concentration of oxygen is stressful for many marine taxa (Vaquer-Sunyer and Duarte, 2008), and we have defined this
concentration as the upper boundary of an “oxygen limited zone” (OLZ) (Stewart et al., 2013). C) Temporal relationship between temperature and dissolved oxygen
concentration at 150 m depth based on daily means of measurements carried out on the same day in a given year. Periods of the sine-wave fits are 362.3 ± 2.8 days
(temperature) and 364.2 ± 4.8 days (oxygen).

6
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

warm season, because oxygen concentration varies greatly over the

course of a year, and the seasonality of temperature and oxygen at depth
are not the same (Fig. 2). A seasonal average was computed by fitting a
sine function to the daily means of CTD data over all years for the depth
at which oxygen was equal to 60 µmol/kg (DOx60; dashed curve in
Fig. 6Ai) or 20 µmol/kg (DOx20; Fig. 6Aii), and the deviations from this
sine wave were then fit with linear regressions covering the same three
time periods as those used for temperature (Fig. 6B). Slopes of fits for
DOx60 for 2015–2020 and 2006–2020 were highly significant, as was
the fit for DOx20 for 2006–2020 (Table 3). No fits to the 2006–2015
period were statistically significant at p < 0.05.
A similar outcome was evident when daily-means data without
adjusting for seasonality were used (not illustrated, Table 3) or with
“raw” data from all individual casts (Fig. 6C; Table 3). In all cases, a
significant shoaling of the upper boundaries of both the OLZ (60 mmol/
kg O2) and OMZ (20 mmol/kg O2) occurred between 2006 and 2020,
with most of that change occurring after 2015. Based on the most con­
servative estimate (deviations from seasonal fit), the shoaling of these
boundaries over 14 years would amount to ~ 60 m for DOx20 and 30 m
Fig. 3. Seasonal variation in water column properties in relation to jumbo for DOx60. Analysis of raw data suggests values of ~ 80 m and 60 m,
squid (Dosidicus gigas) landings in the Santa Rosalia area. Depth of the 22 ◦ C respectively.
isotherm (D22=○) increases during the May-June transition to the warm period Changes in dissolved oxygen concentration at a specific depth (e.g.,
and reaches a maximum in late July. The depth of the 60 µmol/kg O2 isopleth
200 m) were analyzed in the same manner as described above (Fig. S1A,
(solid red line) increases to a peak during this same period and then shoals
B, Supplemental Material). This yielded a significant decreasing slope
during the warm period to a minimum in November. Squid landings (mean ±
SEM) (▴) rise to a maximum around the time D22 reaches 50 m and then fall for the deviation in dissolved oxygen concentration from the fitted
throughout the warm season to an annual low in December-January. D22 (solid seasonal average for the periods 2015–2020 (-1.81 ± 0.78 µmolkg-1yr− 1,
black line) and DOx60 (red line) were smoothed with an 8-point boxcar p = 0.023) and 2006–2020 (-0.98 ± 0.27 µmolkg-1yr− 1, p = 0.0007).
method. Environmental data represent the period of 2006–2020. Squid landings Slopes fit to daily means data were also significant for 2015–2020 (-3.44
are from 2001 to 2008. ± 1.03 µmolkg-1yr− 1, p = 0.001) and 2006–2020 (-1.72 ± 0.34 µmolkg-
1 − 1
yr , p = 0.0007) (Fig. S1C). Slopes fit to the 2006–2015 period were
3.3. Long-term variation in hydrographic properties – Dissolved oxygen not significant.
Considered in aggregate, our analyses of changes in dissolved oxygen
Decreasing oxygen concentration in the water column during at depth reveal very large seasonal changes, with the upper boundaries
2006–2020 was analyzed in a similar manner to that used for temper­ of the OMZ (DOx20) and OLZ (DOx60) varying by ~ 100 m, as well as a
ature. Analysis of oxygen included data for all year and not just the long-term decrease in dissolved oxygen that corresponds to significant
shoaling of DOx20 and DOx60.

Fig. 4. History of hydrographic changes during

the warm season (July 3 – October 31) from 2006
through 2020 in the study area off Santa Rosalia,
Mexico. A) Temperature in the water column
increased at all depths from 2006 to 2015, and
then began an irregular decrease. Strong El Niño
events occurred in 2009–2010 and 2015–2016.
2014 was an unusually warm year. B) Dissolved
oxygen concentration in the water column shows
a slow decrease during most of the monitoring
period. The upper boundary of the oxygen mini­
mum zone (OMZ) is 20 µmol/kg; the upper
boundary of the oxygen limited zone (OLZ) is 60
µmol/kg.

7
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Fig. 5. Subsurface warming and cooling trends during the warm season between 2006 and 2015 and 2015–2020, respectively, in the Santa Rosalia area. A) Daily
means of CTD data (July 1 – October 31) were fit with linear regressions to three time periods: 2006–2020 (dashed lines) and 2006–2015 and 2015–2020 (solid
lines). 2015 data were included in both of the shorter segments. B) Slope of fits for the periods 2006–2015 (●) and 2015–2020 (○) (means ± SD). C) Temperature
changes at each depth given by the fitted slopes for the two time periods analyzed. Diamonds indicate the net temperature change between 2006 and 2020 for depths
where values from both periods are significant (p < 0.05).

Table 2
Summary of linear fits to plots of temperature (daily means, July-October) at specific depths for three periods: 2006–2015, 2015–2020, and 2006–2020. Values of slope
(˚C yr− 1) are means ± SD, r2 is the value for the fit, and the equivalent p value was calculated using an online tool (https://www.socscistatistics.
com/pvalues/pearsondistribution.aspx; accessed 06/23/2022). Values for p < 0.05 are indicated by bold type.
Depth 2006–2015 2015–2020 2006–2020
(m) 2 2
slope n r p slope n r p slope n r2 p

2 − 0.027 ± 0.101 27 0.003 0.788 − 0.378 ± 0.147 39 0.151 0.014 − 0.139 ± 0.061 55 0.089 0.027
5 0.008 ± 0.087 28 0.0004 0.924 − 0.371 ± 0.155 39 0.135 0.022 − 0.113 ± 0.059 56 0.063 0.062
10 − 0.004 ± 0.080 28 0.0001 0.961 − 0.349 ± 0.166 39 0.105 0.045 − 0.109 ± 0.062 56 0.055 0.083
25 0.189 ± 0.091 28 0.144 0.047 − 0.302 ± 0.170 39 0.079 0.084 − 0.004 ± 0.068 56 0.00005 0.960
50 0.348 ± 0.068 27 0.510 0.00003 − 0.384 ± 0.144 39 0.161 0.0116 0.047 ± 0.062 55 0.011 0.452
75 0.348 ± 0.067 27 0.520 0.00002 − 0.402 ± 0.139 39 0.185 0.006 0.068 ± 0.060 55 0.023 0.264
100 0.323 ± 0.072 27 0.453 0.0001 − 0.373 ± 0.106 39 0.251 0.001 0.075 ± 0.052 55 0.038 0.155
125 0.238 ± 0.051 27 0.463 0.0001 − 0.310 ± 0.065 39 0.382 0.00003 0.048 ± 0.036 55 0.032 0.190
150 0.156 ± 0.042 27 0.360 0.0009 − 0.248 ± 0.050 39 0.397 0.00002 0.025 ± 0.028 55 0.014 0.387
175 0.108 ± 0.033 27 0.300 0.003 − 0.201 ± 0.040 39 0.403 0.00001 0.004 ± 0.022 55 0.0006 0.860
200 0.075 ± 0.032 24 0.204 0.027 − 0.189 ± 0.039 39 0.392 0.00002 − 0.017 ± 0.021 52 0.013 0.427
225 0.055 ± 0.032 21 0.136 0.0999 − 0.198 ± 0.039 36 0.427 0.00002 − 0.027 ± 0.020 48 0.039 0.180
250 0.034 ± 0.033 24 0.048 0.305 − 0.165 ± 0.046 37 0.269 0.001 − 0.032 ± 0.021 50 0.045 0.139
275 0.008 ± 0.035 24 0.002 0.834 − 0.108 ± 0.051 36 0.117 0.042 − 0.034 ± 0.021 49 0.051 0.119
300 0.009 ± 0.030 22 0.004 0.769 − 0.104 ± 0.048 34 0.126 0.039 − 0.033 ± 0.020 46 0.056 0.113

8
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Fig. 6. Seasonal and long-term variation in oxygen at depth. A) Daily averages of CTD data obtained on the same day of the year. Data for all years (2006–2020) were
fit with a sine function (dashed wave) as described in the legend to Table 3. Solid irregular curves represent an 18-point boxcar smoothing. Ai) DOx60; n = 146). Aii)
DOx20 (n = 122). B) Deviations of daily DOx60 (Bi) and DOx20 (Bii) data from the corresponding sine fits were fit with linear regressions for the entire monitoring
period (2006–2020; dashed line) and for years 2006–2015 and 2015–2020 (solid lines). Slopes of fits over all the years are − 2.2 myr− 1 for DOx60 (p < 0.001) and
− 4.4 myr− 1 for DOx20 (p < 0.00001) (See also Table 3). C) Linear fits to raw DOx60 (Ci; n = 707) and DOx20 (Cii; n = 420) during 2006–2020 give slopes of − 4.2
myr− 1 (p < 0.00001) for Dox60 and − 5.8 myr− 1 (p < 0.00001) for DOx20 (dashed lines) (See also Table 3). Note that a negative slope in panels B and C appears to be
positive, because of the reversed vertical depth axis traditionally used in oceanographic plots of this sort.

3.4. Long-term relationship between temperature at depth and size of significant (Table 4). It appears that D22 was > 50 m for much of the time
jumbo squid between 2010 through 2019 and reached depths of 100 m or greater.
This pattern of change in D22 shows a striking relationship to a
As discussed previously, a seawater temperature of 22 ◦ C appears to history of the size (ML) of mature jumbo squid (Stages IV and V) in the
mark a threshold, above which large jumbo squid may experience signif­ Gulf of California, primarily the Guaymas Basin (Fig. 8A), and com­
icant physiological stress. Long-term variation in the depth of the 22 ◦ C mercial landings in the same area (Fig. 8B). From its beginning in the
isotherm (D22) is therefore of interest. An assessment of change in this early 1990′ s, this fishery was dominated by squid that matured at a large
feature was made through analyses of both daily means (Fig. 7A) and raw body size (>60 cm ML) in 1–1.5 years (Hoving et al., 2013), and land­
(individual casts) CTD data (Fig. 7B) recorded during the July-October ings in the state of Baja California Sur averaged about 35,000 metric tons
warm season. Both approaches reveal an increase in D22 from ~ 35 m in per year between 1995 and 2008. A major decrease in landings occurred
2006 to an average maximum of ~ 75 m in 2015 followed by a return to ~ in 1998 and was accompanied by a proportionately large decrease in
50 m in 2020. Linear-regression slopes fit to these time periods are highly size at maturity of individual squid, with terminal spawning occurring at

9
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Table 3
Summary of linear fits to DOx60 and DOx20 data for the periods of 2006–2015, 2015–2020, and 2006–2020. Deviations from seasonal fit refers to the difference
between observed daily means from the calculated seasonal value for the specified day (t) based on the following fit parameters (fit to 3 years of concatenated DOx20
and DOx60 daily means data) with the period forced to equal 365 days. DOx20 = 287.71 + 39.002 * sin(0.017213 * t – 1.5491); DOx60 = 166.64 + 48.098 * sin
(0.017213 * t − 1.497). Daily means are data that have not been adjusted for seasonality. Raw data are from individual casts and have not been subjected to any
averaging or any other statistical procedures, although they were processed as described in Methods.
Data Fit 2006–2015 2015–2020 2006–2020

Slope (m/ n r2 p Slope (m/yr) n r2 p Slope (m/yr) n r2 p

yr)

DOx60 deviation from seasonal − 0.49 ± 66 0.001 0.781 − 6.47 ± 2.33 92 0.079 0.0067 − 2.20 ± 0.80 146 0.051 0.0063
fit 1.76
DOx20 deviation from seasonal − 3.31 ± 52 0.047 0.124 − 4.15 ± 2.63 82 0.030 0.1186 − 4.45 ± 122 0.157 0.00001
fit 2.12 0.942
Dox60 daily means − 0.71 ± 66 0.002 0.753 − 11.15 ± 92 0.134 0.0003 − 2.91 ± 1.02 146 0.054 0.005
2.23 2.99
DOx20 daily means − 1.44 ± 52 0.007 0.544 − 9.29 ± 2.97 82 0.109 0.0025 − 5.20 ± 1.09 122 0.160 < 0.00001
2.35
DOx60 raw data − 0.71 ± 161 0.002 0.546 − 8.46 ± 1.23 598 0.074 < 0.00001 − 4.17 ± 0.55 707 0.076 < 0.00001
1.16
DOx20 raw data − 1.16 ± 112 0.005 0.442 − 6.45 ± 1.65 338 0.044 0.0001 − 5.79 ± 0.67 420 0.146 < 0.00001
1.50

an age of ~ 0.5 yr. These phenomena were associated with a strong El

Niño event in 1997–1998 (shaded boxes in Fig. 8), but both landings and
size at maturity recovered within several years. Landings of large squid
continued until another strong El Niño in 2009–2010 when the pattern
of 1997–1998 was repeated – landings crashed, and size of squid at
maturity decreased to ~ 20 cm ML. CTD data for the Guaymas Basin are
not available for years prior to 2006, but the average warm-season D22
during 2006–2010 appears to have been < 50 m (Fig. 7 and filled-in blue
area of curve in Fig. 8).
In the years following El Niño 2009–2010, both squid size and
landings partially recovered, but average D22 remained > 50 m during
this period. In August 2014, squid were landed in Santa Rosalia that
exceeded the historic mean ML prior to 2010 (62.5 cm, excluding 1998
and 1999), but 2014 was an anomalously warm year, with D22 reaching
its maximum depth. This depression of the isotherm was associated with
another shift to the small size-at-maturity phenotype, and all mature
squid sampled in mid-April 2015 were small (20.7 ± 1.8 cm ML (mean
± SD), n = 18, range = 18.5 – 26 cm). These squid were ~ 0.5 yr old
(Hoving et al., 2013; H.-J. Hoving and R. Schwarz, pers. comm., Kiel,
Germany) and would therefore have hatched at the end of the 2014
warm season, well before the start of El Niño 2015–2016. (NOAA, 2015).
All mature squid sampled in the Guaymas Basin since April 2015 have
been of the small phenotype (Gilly et al., 2020). During this period
average, D22 remained > 50 m but has been decreasing steadily, in
parallel with the cooling trend at depth (Fig. 5A).

3.5. Seasonal and interannual hydrographic variation in the southern

Gulf of California

Seasonality of temperature change in the southern Gulf of California

(Fig. 9) is even more marked than that in the Guaymas Basin, with the
warm season developing rapidly in June and ending abruptly in
November (Fig. 9Ai). Oxygen at depths > 100 m increases during the
warm season, corresponding to a depression of the upper boundaries of
Fig. 7. History of D22 during the warm season (July 1-October 31) between
the OLZ (DOx60) and OMZ (DOx20) (Fig. 9Aii). Salinity in the upper
2006 and 2020. A) Daily averages of D22 values (○) were fit with a linear
100 m decreases substantially during the warm season, with tropical
regression for the periods of 2006–2015, 2015–2020 (solid lines). The slopes
are significant (p < 0.01) See Table 4 for fit parameters. Dotted line represents surface water (i.e., temperature > 18 ⁰C and salinity < 35.0; Lavín and
fit for 2006–2020; the slope is not significant. Solid points represent a 16-point Marinone, 2003) appearing in the upper 100 m of the water column
boxcar smoothing. B) D22 raw data from individual CTD casts were fit and (Fig. 9Aiii). This intrusion of tropical surface water influences water
plotted in the same manner. Slopes of fits for all three time periods are highly column properties throughout the southern Gulf of California as well as
significant (p < 0.01). See Table 4 for fit parameters. the Guaymas Basin (Frawley et al., 2019a).

10
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Fig. 8. History of average jumbo squid mantle

length at maturity (A) and commercial landings
(B) in relation to D22 changes in the monitored
area off Santa Rosalia. Strong El Niño events
(1997–1998, 2009–2010, 2015–2016) are indi­
cated by shaded bars. A) Size at maturity is based
on 1057 mature individual squid sampled be­
tween 2005 and 2019 (filled symbols) and on
values taken from the literature for years
1995–2004 (open circles) (Markaida, 2006; Baz­
zino et al., 2007). D22 is the boxcar-smoothed
time course plotted in Fig. 7A,B). Squid land­
ings in Santa Rosalia (filled symbols for
2001–2019) and the state of BCS (open symbols)
are based on Mexican government sources as
described in Methods.

Long-term changes in the southern gulf region have also occurred Analysis was carried out to quantify the subsurface warm-season
since 2010. A warm-season (June 15 – Nov. 30) subsurface warming temperature changes in the southern Gulf of California region in a
trend in the upper 150 m appeared to begin around 2011–2012 and manner similar to that used for Santa Rosalia, the only difference being
reached a peak in 2015 (Fig. 9Bi). This trend and the subsequent cooling the regions fitted with linear regressions over 2010–2015 and
are qualitatively similar to the patterns described above for the Santa 2015–2018, due to the lack of data outside this period. Essentially all
Rosalia area. In this southern region, dissolved oxygen concentration at CTD casts were made on different days, or at substantially different lo­
depth increased during the 2010–2015 warming phase, with the depth cations if on the same day, so daily means were not computed for this
of the OMZ upper boundary (DOx20) falling from 150 to 250 m and then analysis. Data and fits for 5, 50, 100, and 300 m depths show the same
decreasing to < 100 m during the cooling phase after 2016 (Fig. 9Bii). pattern of warming (2010–2015) and subsequent cooling (2015–2018)
Salinity during the 2012–2016 period also decreased in the upper 100 m as seen in Santa Rosalia (Fig. 10A), but the slopes are considerably
and then increased again (Fig. 9Biii). greater in the southern gulf (Fig. 10Bi) – maxima of + 1.68 ◦ Cyr− 1 and

Table 4
Summary of linear fits to D22 data (daily means and raw data for July – October) for the three periods 2006–2015, 2015–2020, and 2006–2020.
Data Fit 2006–2015 2015–2020 2006–2020

Slope (myr¡1) n r2 p Slope (myr¡1) n r2 p Slope (myr¡1) n r2 p

D22 daily means 3.64 ± 0.84 27 0.429 0.0002 − 4.03 ± 1.72 37 0.136 0.0249 0.72 ± 0.71 53 0.018 0.340

D22 raw data 3.68 ± 0.60 99 0.281 0.0048 − 5.42 ± 0.76 243 0.173 0.0069 − 1.14 ± 0.42 291 0.024 0.0075

11
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Fig. 9. Seasonal (left column) and interannual (right column) changes in hydrographic features of the southern Gulf of California (rectangular outlines in maps at
bottom). A) Seasonal changes. Ai) Temperature, Aii) dissolved oxygen, and Aiii) salinity all show large changes in the warm season (June 15 – November 30) due to
the seasonal northward intrusion of tropical surface water. B) Interannual changes. Bi) Temperature shows an increase in depth for thermoclines of 15 to 25 ◦ C
between 2010 and 2015 followed by a decrease. Bii) Dissolved oxygen concentration increased at depth during 2010–2015 and then decreased, in parallel with the
illustrated temperature changes. Biii) Salinity in the upper 100 m also decreased and subsequently increased during the same period.

12
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Fig. 10. Subsurface warming and cooling trends during the warm season (June 15 – November 30) between 2010 and 2015 and 2015–2028, respectively, in the
Southern Gulf of California. A) Data from individual CTD casts were fit with linear regressions over the time periods of 2010–2015 and 2015–2018 (solid line
segments). Dashed lines indicate values in October 2001. Bi) Slope of fits for the periods 2010–2015 (●) and 2015–2018 (○) (means ± SD). Bii) Temperature changes
at each depth as given by the fitted slopes for the two time periods analyzed. Diamonds indicate the net temperature change between 2010 and 2018 for depths were
values form both time periods are significant (p < 0.05). C) History of D22, DOx60, and DOX20 during the warm season. Straight line segments represent fits of linear
regressions to the time periods of 2010–2015 and 2015–2018.

–3.96 ◦ Cyr− 1 at 50 m depth. These values are 5 and 10 times greater than 3.6. Changes in Gulf of California fisheries
the respective maxima in Santa Rosalia (+0.35 ◦ Cyr-1at 50 m depth and
− 0.39 ◦ Cyr− 1 at 75 m depth; Fig. 5B). Overall temperature changes in 3.6.1. Large teleost fishes − sport versus commercial fisheries
the southern Gulf of California region during the warming and cooling Many fisheries in the Gulf of California, in addition to that for jumbo
phase were also much larger, reaching nearly ± 10 ◦ C (Fig. 10Bii). squid, underwent major changes during the period of subsurface
Changes in the depth of the 22 ◦ C isotherm (D22) in the warm season warming (2010–2015) and subsequent cooling (2015–2018) docu­
during 2010–2018 (Fig. 10C) were quite similar to those in Santa mented throughout the region. Two independent sources of sport-fishery
Rosalia. Fits to D22 suggest that “average” D22 exceeded 50 m in 2013, data from the southernmost part of the Gulf of California showed highly
reached a maximum depth of ~ 70 m in the fall of 2015 and then significant declines on a catch-per-trip basis for 4 species (yellowfin
decreased to < 50 m in 2016. Subsurface oxygen changes in this tuna, dorado, sierra, and black skipjack; see Table 1) that make up about
southern region during 2010–2018 were much more dramatic than 50 % of all sport fish caught by both sources from which data were
those in Santa Rosalia, and the large increase in both DOx60 and DOx20 collected (Fig. 11A). Catch per unit effort (fish per charter trip)
over 2010–2015 was not present in Santa Rosalia. decreased ~ 50 % from pre-2010 levels by the time of El Niño
2015–2016 and remained low afterwards, despite no significant change
in the number of charter trips (open circles in Fig. 11A). Individual

13
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Fig. 11. History of selected sport and commercial fisheries in Gulf of California. A) Sport landings from two sources in the San José del Cabo area (Fig. 1). Open
symbols represent total charter trips for Gordo Banks Pangas (○) and John Snow (⋄, range = 52–74). Filled symbols represent the aggregate number of yellowfin tuna,
dorado, black skipjack, and sierra (see Table 1) caught per charter trip for Gordo Banks Pangas (●) and J. Snow (◆). Fitted slopes (solid lines) are − 0.26 ± 0.06 fish/
trip⋅yr− 1 for Gordo Banks and − 0.21 ± 0.06 fish/trip⋅yr− 1 for J. Snow; both values are highly significant (p < 0.001). B) Average values of catch per charter trip for
individual species from the two sources: yellowfin tuna (●), dorado (∇), black skipjack (◆), and sierra (▴). Slopes of fits are given in Table 1. C) Commercial
landings for the same four species (summed) from the four states bordering the Gulf of California (○) show an increasing slope (1264 ± 588 ton⋅yr− 1, p = 0.051), in
contrast to a declining slope in catch per trip for sport landings of the same species (● = averaged data of panel A; slope = -0.236 ± 0.033 fish/trip yr− 1, p = 0.0004).
D) Commercial landings for the individual four species (same symbols as in panel B) show different patterns than those found for sport fishing (panel B). E) Average
sport landings for pargos (●) showed large oscillations, in comparison to commercial landings (□) that showed a more regular increase (slope = 445.8 ± 93.9 tons
yr− 1, p = 0.0004). F) Commercial landings for all six species of small pelagics (○) rose to a peak in 2008–2009. Landings for sardines (Δ) and the 5 non-sardine species
(∇, see Methods) changed in a roughly reciprocal manner from 2007 to 2016, with sardine landings then rising. The rising slope for non-sardine species is 16,643 ±
4,820 ton⋅yr− 1, p = 0.004). Average sport landings per charter for the 18 groups monitored by both sources (●) declined steadily after 2007 (slope = -0.305 ± 0.052
fish/trip⋅yr− 1, p < 0.0001).

species showed a similar overall pattern of decline, although fluctua­ patterns of change versus those for sport fishing (Fig. 11D). Commercial
tions in yellowfin tuna and black skipjack were larger than those for landings for tuna showed an increasing trend with unusually high
dorado or sierra (Fig. 11B). landings in 2014 and low numbers in 2017. Sierra landings also
These same four species were also landed by commercial fishing increased steadily from 2005 to 2020 but showed no large fluctuations.
efforts in the states bordering the Gulf of California (Baja California, Baja Black skipjack landings decreased prior to El Niño 2009–10 and then
California Sur, Sonora, and Sinaloa). Total landings of these four species remained low. Dorado landings increased prior to 2009–2010 and then
in the Gulf of California increased between 2005 and 2020, opposite to rapidly fell to a low level, but landings are very small compared to the
the declining trend for sport-caught fish of the same species (Fig. 11C). other fish species taken. It should be noted that much of the catch,
Analysis of individual commercial species reveals quite different particularly for yellowfin tuna and black skipjack, is probably taken by

14
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

purse-seine vessels operating in waters well outside the Gulf of (Clarke and Paliza, 2001; Jaquet and Gendron, 2002) – but in 2010,
California. squid of the large phenotype started to disappear from the Gulf of Cal­
In general, commercial landings for these individual species do not ifornia and were completely gone by 2015. In other cases, particularly
show an overall pattern of decline during 2005–2020, but reasonably those of declining sport and commercial fishery landings for both
strong correlations exist for some species between commercial landings pelagic and demersal teleost fishes, any direct connection to the sub­
and sport catch-per-trip values (Fig. S2, Supplemental Material). Sig­ surface warming trend or other climatic drivers is much less clear.
nificant negative correlations were evident for yellowfin tuna (Fig. S2A, Nonetheless, a general picture emerges of overall decline in pelagic
p = 0.035) and sierra (Fig. S2B, p = 0.00005), whereas a positive cor­ ecosystems of the southern half of the Gulf of California during a time of
relation was found for dorado (Fig. S2C, p = 0.030). A suggestive pos­ major changes in features of the water column to depths of several
itive correlation for skipjack was not significant (data not shown, p = hundred meters.
0.144).
Pargos (mixed snappers), the third most frequently landed sport 4.1. Environmental changes
group (Table 1), did not show a significant declining slope in catch per
charter trip but rather a decreasing trend through 2013 followed by Several important trends in hydrographic properties of the Guaymas
large but transient increase (solid circles, Fig. 11E). Commercial land­ Basin near Santa Rosalia have been documented in the present study.
ings showed a significant (p = 0.0004) rising slope over the entire First, strong seasonal variation in temperature and dissolved oxygen
2005–2020 period (open circles, Fig. 11E), although a rapid rise and fall concentration at depth occurs, with the depth of the OMZ upper
after 2013 is evident, similar to that seen with sport fish. This pattern boundary (DOx20) varying over a 75 m range, with a maximum depth
was seen in data form both sport fishing sources (Fig. S3). A suggestive of ~ 325 m in June and a minimum of ~ 250 m in January. This seasonal
positive correlation between sport and commercial data sets for pargo cycle does not match the cycle for temperature that has a peak during
was not significant (data not illustrated, p = 0.11). August-September. Second, oxygen at depth decreased between 2006
Histories of catch per trip for all 19 types of sport fish (Table 1) are and 2020, with DOx20 shoaling at an average rate of ~ 4 myr− 1 for the
given for both data sources (and the average) in Fig. S3. Although no period 2006–2020 and DOx60 (OLZ upper boundary) shoaling at ~ 2
obvious difference in rate of decline for the ensemble group was evident myr− 1. These rates were considerably higher if computed for 2015–2020
for the warming (2010–2015) and cooling (2015–2018) periods in this (6.5 myr− 1 for DOx20 and 4 myr− 1 for DOx60). Third, temperature at
region (Fig. 11F), individual species do show some hints of this. Yel­ depth increased significantly in the 75–175 m depth range during the
lowfin tuna, the most frequently landed fish (Table 1) increased from warm season (July-October) between 2006 and 2015, with a maximum
2011 to 2017 and then decreased rapidly. Dorado, the second most temperature increase of ~ 3 ◦ C at 100 m depth. Finally, significant
frequently reported species, showed a reciprocal pattern – decreasing cooling occurred between 2015 and 2020, resulting in a net increase of
during 2012–2016 and then increasing through 2019. Pargos, with the ~ 1 ◦ C at 100 m in 2020 relative to 2006. A previous description of the
third largest landings, decreased from 2011 to 2014, increased to a peak warming trend in the Guaymas Basin based on data through 2017 did
in 2016 and then decreased again. Sierra, fourth in total landings, not detect this important cooling trend, but its onset is apparent in the
showed a slow decline up to 2013–-2014 and then a rapid drop between published data (see Fig. 7a of Frawley et al, 2019a).
2015 and 2016. Qualitatively similar seasonal and interannual trends in subsurface
temperature also occurred in the southern Gulf of California. In this case
3.6.2. Small pelagic fishes the maximum (warm season) temperature change between 2010 and
Commercial landings of small pelagic fishes (California sardine and 2015 reached ~ 10 ◦ C, and a subsequent cooling trend resulted in a net
five “non-sardine” species, see Materials and Methods) showed large change of ~ 1 ◦ C at 50 m in 2018 relative to 2010. Variation in depth of
changes during the years of interest. Sardine landings increased about the upper boundary of both the OMZ and OLZ were more dramatic than
threefold in 2008 and 2009 before decreasing drastically in 2010, falling in Santa Rosalia, both seasonally and over the decade 2008–2018. These
nearly to zero in 2014 (Fig. 11F). Landings of the five other species changes paralleled those in temperature, but they were opposite to the
showed a reciprocal trend – decreasing to a historically low level prior to direction expected for changes in oxygen solubility with temperature, i.
El Niño 2009–2010 and then increasing rapidly as sardine landings fell e., both temperature and dissolved oxygen concertation at depth
between 2010 and 2013. A significant negative correlation was found increased during the 2010–2015 warming period.
between landings of non-sardine species vs sardine over the years
2005–2020 (Fig. S2D, Supplemental Material; p = 0.006). 4.1.1. Warming in the Gulf of California and the California Current System
Average catch-per-trip data for the 18 types of sport fish tabulated Seasonal and interannual variation in subsurface temperature in the
(Blue/black marlin were not included, See Table 1) showed a steady southern Gulf of California and the Guaymas Basin (Santa Rosalia) ap­
decline starting in 2007 (solid circles in Fig. 11F), and a negative cor­ pears to be due to variation in the severity of intrusions of tropical
relation of these data was found with commercial landings of the five surface water into the Gulf of California from regions to the south during
non-sardine species (Fig. S2D, p = 0.016). No correlation between sport- the warm season (Frawley et al., 2019a). This process also transports
fish landings and commercial landings of sardine or the total for all six zooplankton of tropical affinity northward and typically results in a
small pelagic species was evident (data not illustrated). decreased biomass of the zooplankton assemblage (Brinton and Town­
send, 1980; Brinton et al., 1986; García-Fernández et al., this volume).
4. Discussion The influence of this source diminishes northward with distance from
the mouth of the Gulf of California (Portela et al., 2016). Differences
This study demonstrates environmental and ecological (including between environmental trends reported for Santa Rosalia and the
fisheries) changes in the Gulf of California that have occurred over the southern Gulf of California can potentially be attributed to mesoscale
last 10–20 years. In some cases, particularly that of large jumbo squid oceanographic processes (i.e. filament induced transport) allowing for
(Dosidicus gigas) in the Guaymas Basin region, relatively clear seasonal the seasonal exchange of water masses between Santa Rosalia and the
and interannual connections can be made between climatic change in turbulent and well-mixed Midriff Islands region (Navarro-Olache et al.,
subsurface features of the water column, particularly depth of the 22 ◦ 2004; Frawley et al., 2019a).
isotherm, mantle length (and age) of adult squid at maturity, and the Years covered in the present study included two strong El Niño
state of the commercial squid fishery. A closely related ecological ca­ events (2009–2010 and 2015–2016) that contributed to warming at
sualty appears to be the sperm whale (Gilly et al., 2020), an apex depth, both in the southern Gulf of California and the Pacific Ocean.
predator that prefers to forage on large jumbo squid if they are available Such events are also associated with increased poleward advection of

15
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

tropical water into the gulf and along the Pacific coast of the Baja Cal­ (Figs. S1, Fig. 6) and the wider Guaymas Basin (unpublished data), has a
ifornia peninsula (Durazo et al., 2017). El Niño 2015–2016 led to a connection to nitrogenous agricultural runoff into the Gulf of California
variety of ecological impacts along the Pacific coast, including decreased from mainland Mexico, particularly the Yaqui Valley, south of Guaymas.
nutrients in the upper water column and large changes in the phyto­ Fertilizer use in the state of Sonora has increased greatly since the
plankton community composition (González-Silvera et al., 2020), 1960′ s, and a great deal of nitrogen (primarily ammonia) passes
decreased primary productivity (Gómez-Ocampo et al., 2017), and an seasonally into the Guaymas Basin through an elaborate system of canals
altered biogeographic distribution of pelagic red crabs (Pleuroncodes that drain the fields (Ahrens et al., 2008). This runoff has been impli­
planipes) (Cimino et al., 2021). In the Gulf of California, El Niños cated with major algal blooms in the Guaymas Basin and a significant
2009–2011 and 2015–2016 have been associated with a decreased contribution to primary productivity in this region (Beman et al., 2005).
biomass of micronekton that occupy a critical midtrophic position in the However, a recent, thorough review of nitrogenous inputs into the entire
pelagic ecosystem (Portner et al., 2022). Gulf of California indicates that agriculture (and shrimp-farming) con­
An important feature common to the Gulf of California and the Pa­ tributes only a small fraction of the total entering from the Pacific Ocean
cific coast of the Baja California peninsula was the anomalously warm (Páez-Osuna et al., 2017). This important aspect of productivity, and
year of 2014. This feature was demonstrated in the present study for the potentially OMZ shoaling, particularly in the Guaymas Basin requires
Gulf of California (see also Martínez-Soler et al., 2021) and was also more attention. This is of general concern, because climatic warming is
reported for the Pacific Ocean over the same latitudinal range (Gilly affecting OMZ shoaling in both the Eastern Tropical Pacific Ocean
et al., 2020). Further north, in the Southern California Current System (Stramma et al., 2008) and the California Current System (Howard et al.,
(Ohman, 2018), anomalous warming during 2014 (Rudnick et al., 2017) 2020).
was associated with a variety of ecological effects (Cavole et al., 2016;
Gómez-Ocampo et al., 2017; Lilly and Ohman, 2018, 2021; Morrow 4.2. Ecological changes – Jumbo squid
et al., 2018).
All of these anomalies occurred largely contemporaneously with a During the years prior to 2010, jumbo squid in the Guaymas Basin
major warming anomaly during 2013–2014 in the northeast Pacific migrated between the Baja California peninsula (warm season) and
Ocean that has come to be known as “The Blob”, a phenomenon caused Sonora (cool season) (Markaida et al., 2005), and the flourishing fish­
by a decrease in mixed layer depth and a reduced rate of sea-surface eries in Santa Rosalia and Guaymas (Sonora) were therefore ~ 6 months
cooling (Bond et al., 2015). out of phase. Most importantly, this migration allowed the squid to
Thus, the Blob and the 2014 anomalies in the Southern California forage year-round in productive areas, which supported an ability to live
System (and the Gulf of California) do not share a common mechanism for up to 1.5 years or more and achieve a large body size before
and are thought to be independent (Ohman, 2018; Martínez-Soler et al., maturing, spawning, and dying. This pattern was disrupted by a strong
2021). How large-scale global atmospheric drivers may be influencing El Niño in 2009–2010 that led to precocious spawning at very small size,
marine heat waves in the Gulf of California, in southern California, and representing a switch to the phenotype characteristic of the species in
elsewhere in the northeastern Pacific Ocean, and how these processes the tropical, equatorial portion of its large range (Hoving et al., 2013).
interact with the El Niño-Southern Oscillation (ENSO) cycle remain This same phenotypic switch occurred in conjunction with the strong El
critical questions (Capotondi et al., 2015; Cavole et al., 2016; Gente­ Niño 1997–1998, and full recovery of the large size-at-maturity
mann et al., 2017; Cornwall 2019). In particular, the Gulf of California phenotype (and the commercial fishery occurred within several years
may respond differently than the Pacific, because coastal trapped Kelvin (Fig. 8), aided by a strong La Niña phase during 1999–2002. Incipient
waves that propagate northward along the eastern boundary during El recovery of body size and the fishery was evident after El Niño
Niño events get trapped in the Gulf of California leading to a much more 2009–2010, but the subsequent subsurface warming trend culminated in
stratified water column that persists for a longer time (Baumgartner and the anomalously warm 2014 and another strong El Niño in 2015–2016.
Christensen, 1985; Lavín and Marinone, 2003; Lluch-Cota et al., 2010). Despite the significant cooling trend since 2015, squid have remained
small, and the fishery has disappeared due to economic constraints
4.1.2. OMZ shoaling in the Gulf of California and elsewhere in the eastern associated with profitability of capturing small squid in commercial
Pacific Ocean quantity.
Long-term changes in dissolved oxygen concentration in the Santa Depth of the 22 ◦ C isotherm (D22) is strongly correlated with the
Rosalia region of the Guaymas Basin can be characterized as a gradual ecological status of jumbo squid in the Gulf of California. On a seasonal
shoaling of the upper boundaries of the OMZ (DOx20) and OLZ (DOx60) basis, historic fishery landings in Santa Rosalia started to decline during
between 2006 and 2020 and do not parallel the warming and cooling August-September when D22 approached or exceeded 50 m. On a long-
trend so obviously as in the southern Gulf of California region. Mecha­ term basis, we propose that if D22 exceeds 50 m for a significant length
nisms behind the large seasonal variation and historic shoaling in Santa of time, either during a strong El Niño or the annual warm season, at
Rosalia are likely related to mesoscale links to the Midriff Islands region least a fraction of the squid (and possibly all) will take on the small,
as discussed above in conjunction with subsurface temperature changes. tropical phenotype in the affected area. Undoubtedly changes in food
Increasing depth of the OMZ from the mouth of the gulf to the northern availability reinforce any direct temperature signal, and prolonged,
edge of the Guaymas Basin suggests that changes in oxygen in the upper subsurface temperature changes are likely to alter the composition and/
water column at the mouth probably have little direct effect on oxygen or biomass of plankton and micronekton communities on which squid
at much greater depths in the Guaymas Basin (Fig. S4). How long-term foraging depends. Several contributions to this volume examine this
shoaling of the OMZ in the Guaymas Basin is related to this phenomenon important question (Portner et al., 2022; García-Fernández et al., this
elsewhere in the Eastern Pacific (Stramma et al., 2010; Gilly et al., 2013) volume).
is unknown, but the rate of oxygen change at 200 m depth over In addition, long-term subsurface thermal changes, as well as pre­
2006–2020 (conservatively, ~ − 1 µmolkg-1yr-1) appears to be sub­ vailing wind patterns, can alter the vertical or geographic location of
stantially greater than that reported for an area in the Pacific Ocean prey in a way that squid cannot forage on a year-round basis to the
immediately offshore of the Baja California peninsula between the pe­ degree necessary to support the large phenotype. This appears to be
riods 1960–1974 and 1990–2008 (Stramma et al., 2008). Unfortunately, exactly what happened after El Niño 2009–2010, when a years-long
data for dissolved oxygen concentration are not available for the Gulf of collapse of winter winds driving upwelling was accompanied by the
California for those earlier periods. subsurface warming trend described in the present study (Robinson
It is possible that the high rate of oxygen decline at 200 m and et al., 2016; Frawley et al., 2019a). In short, the Gulf of California
shoaling of the OMZ upper boundary in the Santa Rosalia region became quite unfavorable to supporting the large phenotype of jumbo

16
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

squid after 2010, and the small phenotype appears to have become found in surface waters of the Gulf of California that are warmer than 24
firmly fixed in the region until the present (2022). ⁰C (Hammann et al., 1998). We are unaware of any data on the thermal
sensitivity of sardines (or the other small pelagics) inhabiting the Gulf of
4.3. Ecological changes - fisheries California, but sardines from the Pacific “southern” stock (Ensenada) are
very sensitive to temperature above 22 ◦ C and actively avoid this con­
Commercial squid landings in the Gulf of California have shown the dition, with death becoming incipient above 25.6 ◦ C (Martínez-Porchas
most dramatic decline during the last decade (Fig. 8) − one that has been et al., 2009, 2011). Thus, the depth of the 22 ◦ C isotherm may be as
attributed primarily to environmental change (Robinson et al., 2016; critical for sardines, and potentially for other small pelagics, as it is for
Frawley et al., 2019a). Squid landings have essentially been zero since jumbo squid (Fig. 8E), and the years 2010–2015 were likely to have been
2015, because small squid cannot be caught in quantities demanded by increasingly unfavorable for sardines in the Gulf of California.
the market using standard jigging methods, and participants have sim­
ply abandoned the fishery. Because of this de facto gear limitation, 4.3.2. Small pelagic fishes and sport fish
which affected both industrial and artisanal fishing efforts, the squid Negative impacts of warming events and fishing pressure on small
fishery in the Gulf of California presents an unusual case in which pelagic fishes have been well documented for nesting elegant terns
overfishing is thought to not have been a major factor in a collapse. This (Thalasseus elegans) on Isla Rasa in the central Gulf of California, where
idea should be qualified by the possibility that an intense increase in seabirds compete with commercial fishing operations for the same small
fishing pressure in the summer of 2014 during the nascent recovery of pelagic fishes (Velarde et al., 2015, 2019). Warming has also been
large squid may have been a factor along with El Niño 2015–2016 in re- implicated in the long-term decline in the California sea lion (Zalophus
establishing the small phenotype. californianus) throughout most of the Gulf of California (Adame et al.
2020). Potential impacts of both warming and commercial fishing
4.3.1. Sardines and other small pelagic fishes pressure might also be expected for predatory fishes that also forage on
Commercial fisheries for small pelagics, historically the largest and small pelagics. Indeed, the reciprocal trends in landings of non-sardine
most valuable fishery in Mexico, also changed radically during this small pelagics (rising trend) and the total landings per charter trip for
warming period. Sardine landings surged in 2008–2009 and then the 18 groups of sport fish analyzed (falling trend) present a troubling
collapsed after El Niño 2009–2010, reaching a level close to zero by scenario (Fig. 11F, Fig. S2E).
2014–2015. Sardines have recovered somewhat, and landings presently It is unlikely that fishing pressure for the targeted species would be a
are close to levels recorded before 2007. Landings of the 5 other small main factor in declining sport catch per unit effort for several reasons.
pelagic species rose while sardine landings were crashing after First, the number of sport charter-trips was relatively constant during
2009–2010, as El Niño probably led to an increased abundance of the monitoring period, but the catch per charter for 10 of the 18 groups
southern species like thread herring, and efforts of the sardine fleet were monitored showed significant (p < 0.05) or marginally significant (0.05
redirected toward these species. Aggregate landings of these non-sardine ≤ p < 0.10) rates of decline. Second, declines were found for both six
species have fallen somewhat from the 2013 peak, but they remain very pelagic and four demersal types of fish, even though demersal species
high at ~ 350,000 tons per year. Taken together, total landings for the would be expected to be more readily overfished by rod-and-reel sport
six small pelagics have approximately doubled between 2005 and 2020. methods. Third, individual commercial landings for four of the top five
How this level of harvest might affect the Gulf of California ecosystem is sport species (yellowfin tuna, dorado, sierra, black skipjack) did not
a critical question (Arreguín-Sánchez et al., 2021). show a consistent pattern of decline or increase. However, these
Sardines are sensitive to environmental thermal variation on a va­ migratory pelagics may not enter the Gulf of California in the same
riety of temporal and spatial scales, with recruitment and landings being numbers every year, depending on broader oceanic conditions and
negatively impacted by El Niño and other warming events in the commercial fishing pressure offshore in the Pacific Ocean.
southern part of its range, including the Gulf of California (Nevárez- A positive correlation between sport catch-per-trip numbers and
Martínez et al., 2001; Galindo-Cortes et al., 2010; Petatán-Ramírez et al., commercial landings for dorado (Fig. S2C), and nominally pargos
2019). To what extent fishing pressure may contribute to fluctuations in (Fig. 11E), suggests that common environmental factors might have
sardine landings has been highly controversial since the crash of Pacific negatively influenced both sport and commercial fishing ladings. Sig­
sardine fishery in the late 1940′ s (Ricketts, 1947; Ricketts, 1948; nificant negative correlations between sport and commercial data sets
Radovich, 1982; McFarlane et al., 2005; Fisher, 2020; Izquierdo-Peña for sierra (Fig. S2B) and yellowfin tuna (Fig. S2A) could be consistent
et al., 2020; Moore, 2021), and effects of fishing and environmental with a direct influence of increasing commercial fishing pressure on a
variation on sardine landings in the Gulf of California are most likely decline in sport fish catch per unit effort, including large-scale com­
comparable (Giron-Nava et al., 2021). Similar analyses for the non- mercial efforts in distant waters outside of the Gulf of California.
sardine species that also provide essential forage for many species of Although significant declines in sportfish landings occurred during
fish, birds, and marine mammals are lacking. the warming trend of 2009–2015 (Table 1), some of the trends started in
It is possible that the subsurface warming trend of 2010–2015 2005. It is thus unclear which environmental changes, if any, might be
impacted sardines (and the other small pelagics) in the Gulf of California responsible for the overall trend of declines for individual types of fish.
in other ways. One possibility is that sardines moved out of the Gulf and Multidecadal declines in a number of demersal grouper and snapper
migrated northward to cooler waters of the Pacific Ocean (Petatán- species in the Gulf of California have been documented using social-
Ramírez et al., 2019). However, a subsurface warming trend was also science techniques and attributed to overfishing (Sáenz-Arroyo et al.,
evident in the Pacific Ocean off the southern half of the Baja California 2005), and fishing pressure on highly migratory pelagics targeted by
Peninsula during this period (Gilly et al., 2020), so unfavorable condi­ both commercial and sport fishing in the gulf may directly lead to
tions in these waters might have pushed migration northward to Punta similar effects on vastly different time scales.
Eugenia or beyond. Another possibility is that sardines formed smaller A troubling possibility emerges that sportfish populations of diverse
schools and/or moved deeper in the water column as warming pro­ species in the southern Gulf of California, like terns on Isla Rasa far to the
gressed, making them less accessible to the fishery (Rubio-Rodríguez north and sea lions at many locations in between, have been adversely
et al., 2018), but effects of school size and density on landings are also affected by a decreased supply of small pelagic forage fish driven by the
controversial. warming trend post-2010 and increased commercial fishing pressure,
Temperatures in the upper water column of the Gulf of California particularly on non-sardine species. This critical issue warrants further
during 2014–2016 appear to have been high enough to have directly investigation. Indeed, regional small-scale fishers in the study region
impacted sardine development and survival. Sardine eggs are rarely firmly believe that the activities of industrial small-pelagic fisheries

17
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

have contributed to their declining catch rates by undermining the efforts should be made to build local research capacity and increase
forage base upon which their livelihoods depend (Frawley et al., 2019b). community participation. Indeed, the future health and sustainability of
the Gulf of California will depend far more on the degree to which local
5. Conclusions individuals and organizations engage in the processes of environmental
monitoring and resource stewardship than on scientific studies doc­
Given the significant sub-surface cooling trend since 2015 in both the umenting decline.
Guaymas Basin and southern gulf region, it is reasonable to ask when
large jumbo squid might reappear in the Gulf of California. Since ther­ Declaration of Competing Interest
mal effects appear to be largely (albeit indirectly) responsible for
maintaining the small phenotype, return of large squid will require re­ The authors declare that they have no known competing financial
covery of the ability of squid to forage for preferred prey in the upper interests or personal relationships that could have appeared to influence
water column at night and the ability to locate adequate food resources the work reported in this paper.
on a year-round basis through migrations within the Gulf of California or
between the gulf and Pacific Ocean. These processes will not instantly Data availability
recover as the thermal landscape returns to a pre-2010 “normal” situa­
tion. Given that it took three La Niña years for recovery of squid size and Data will be made available on request.
the fishery after El Niño 1997–1998, this amount of time should be
expected to represent the perfect situation – but after a major subsurface Acknowledgements
warming trend bracketed by two strong El Niño events, it is clear that
the Gulf is not in a state like that in 1999 given the changes we have We dedicate this paper to the memory of our colleague, Karmina
documented since 2005. Arroyo of Instituto Tecnológico Superior de Mulegé (ITESME) in Santa
It is thus conceivable that the jumbo squid’s shift to the small Rosalia, BCS, Mexico, who passed away during the writing of this paper
phenotype in the Gulf of California will be very long-lasting, or even to which she contributed so much. Her knowledge, enthusiasm, and
permanent. If so, this should be viewed as a poleward migration of the personal warmth will always be remembered.
tropical phenotype and its fixation by decreasing productivity in the We thank Eric Bricston, Gordo Banks Pangas, and Dr. John T. Snow
Gulf – and not an “invasion” of individual small squid from the south. It for providing data on sport fish landings and for their careful record
seems most likely that large squid can give rise to a generation of small keeping; the many fishermen of Santa Rosalia and Guaymas with whom
squid in response to a strong El Niño, and that small squid can then we have worked, the captains and crew of the many vessels that were
persist for many generations if conditions remain unfavorable. A similar used in the Gulf of California during this study; and Pablo Zambrano and
situation may be an ongoing feature of the Humboldt Current System, the Directors and staff of Instituto Tecnológico Superior de Mulegé
where temperature in early development plays a critical role in (ITESME) in Santa Rosalia, BCS, Mexico for their support and hospital­
achieving large size at maturity (Arkhipkin et al., 2015b), and both small ity. We are particularly grateful to Mickey Steward, Leonel Orozco, Olga
and large phenotypes are found in constantly moving areas of variable Zuñiga, and the environmental monitoring team at Minera y Metal­
productivity (Tafur et al., 2010; Argüelles et al., 2008; Keyl et al., 2008, úrgica del Boleo, Santa Rosalia, for supporting and carrying out many of
Keyl et al., 2011). the CTD casts analyzed in this paper. We thank Henk-Jan Hoving and
If large squid do return to the Gulf of California, it seems likely that Richard Schwarz for carrying out statolith-aging studies on small,
recovery will occur incrementally based on a geographic separation of mature jumbo squid captured in summer 2016.
unequal resources within the guff that keep squid in one region locked in We also thank the crews of the R/V El Puma (UNAM), students, and
the small phenotype while permitting squid in another region to live for researchers from the Laboratorio de Ecología de Pesquerías (Instituto de
multiple seasons and achieve large body size. Indeed, such a situation Ciencias del Mar y Limnología-Universidad Nacional Autónoma de
appears to have existed historically in the southern Gulf of California. In México), Centro Interdisciplinario de Ciencias Marinas-Instituto
May 2008, a year before the start of El Niño 2009–2010, we sampled Politécnico Nacional, for recording environmental information and
both large and small phenotypes of mature jumbo squid on the same day collecting zooplankton samples.
near Isla San Ignacio Farallon, about 400 km south of the Guaymas Basin This work was supported by the U.S. National Science Foundation
(Hoving et al., 2013). Coexistence of the two phenotypes in a given (OCE 0526640, EAGER OCE 0850839, OCE 13389739, RAPID IOS
region is likely to be aided by the tendency for larger squid to inhabit 1420693, IOS 1557754); National Geographic Society Committee for
near-shore shelf regions, with smaller squid occupying a more offshore Research and Exploration (7578-04 and 9366-13); Lindblad
blue-water habitat. Cannibalism is probably another factor that would Expeditions-National Geographic Fund (LXII-15); David and Lucille
tend to keep the two phenotypes physically distant. Whether the two Packard Foundation ((2005-2800 and 32708); Tagging of Pacific Pe­
phenotypes could coexist stably in a relatively small body of water like lagics (Census of Marine Life); and Stanford University IntroSems Plus
the Gulf of California is unknown. program. Support from sources in Mexico included Minera y Metal­
From a broader perspective, the apparent ongoing declines in other úrgica del Boleo through SURMAR-ASIMAR (The Ocean Foundation) to
taxa, specifically sea lions (Adame et al., 2020), small pelagics, and a WFG; and PAPIIT, UNAM (IN20066610-3), and CONACYT (152580) to
wide variety of sport fish should raise other concerns about the future of CJR and JG-G. UNAM supported the use of the RV El Puma during twelve
the Gulf of California. In addition to ecological issues, numerous other oceanographic cruises (2005–2017).
societal issues emerged during this study as local resource users strug­
gled to adapt to a new reality of no large jumbo squid. One positive Appendix A. Supplementary material
aspect concerns the increasing role of disparate “citizen science” efforts
in collecting data in the present study. Over the last 20 + years we have Supplementary data to this article can be found online at https://doi.
worked with fishers in the Santa Rosalia community and squid pro­ org/10.1016/j.pocean.2022.102857.
cessors in Guaymas, but we also engaged with Lindblad Expeditions, a
well-known ecotourism company aligned with National Geographic, References
and Minera y Metalúrgica del Boleo, a copper mining corporation
engaged in monitoring efforts in Santa Rosalia, to collect CTD data. Aburto-Oropeza, O., Ezcurra, E., Moxley, J., Sánchez-Rodríguez, A., Mascarenas-
Osorio, I., Sánchez-Ortiz, C., Erisman, B., Ricketts, T., 2015. A framework to assess
Sport-fish data were painstakingly collected by concerned sportsmen. the health of rocky reefs linking geomorphology, community assemblage, and fish
This degree of community involvement is encouraging, and additional biomass. Ecol. Indic. 52, 353–361. https://doi.org/10.1016/j.ecolind.2014.12.006.

18
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Adame, K., Elorriaga-Verplancken, F.R., Beier, E., Acevedo-Whitehouse, K., Pardo, M.A., Clarke, R., Paliza, O., 2001. The food of sperm whales in the southeast Pacific. Mar.
2020. The demographic decline of a sea lion population followed multi-decadal sea Mammal Sci. 17, 427–429. https://doi.org/10.1111/j.1748-7692.2001.tb01287.x.
surface warming. Sci. Rep. 10 (1), 1–15. https://doi.org/10.1038/s41598-020- Cornwall, W., 2019. In hot water. Science 363 (6426), 442–445. https://doi.org/
67534-0. 10.1126/science.363.6426.442.
Ahrens, T.D., Beman, J.M., Harrison, J.A., Jewett, P.K., Matson, P.A., 2008. A synthesis Davis, R.W., Jaquet, N., Gendron, D., Markaida, U., Bazzino, G., Gilly, W., 2007. Diving
of nitrogen transformations and transfers from land to sea in the Yaqui Valley behavior of sperm whales in relation to the behavior of a major prey-species, the
agricultural region of northwest Mexico. Water Resour. Res. 44 W00A05 https://doi. jumbo squid, in the Gulf of California. Mexico. Mar. Ecol. Prog. Ser. 333, 291–302.
org/10.1029/2007WR006661. https://doi.org/10.3354/meps333291.
Alvarez, C., Andraka, S., Anhalzer, G., Morgan, S., 2017. Small pelagics fishery in Sonora, Durazo, R., Castro, R., Miranda, L.E., Delgadillo-Hinojosa, F., Mejía-Trejo, A., 2017.
Gulf of California. MSC Fishery Assesment Report. SCS Global Services, Emeryville, Anomalous hydrographic conditions off the northwestern coast of the Baja California
CA. December 19, 2017.Table 4, p. 29. https://fisheries.msc.org/en/fisheries/small- Peninsula during 2013–2016. Cienc. Mar. 43 (2), 81–92. https://doi.org/10.7773/
pelagics-fishery-in-sonora-gulf-of-california/@@assessments. Accessed 06/23/2022. cm.v43i2.2754.
Argüelles, J., Tafur, R., Taipe, A., Villegas, P., Keyl, F., Dominguez, N., Salazar, M., 2008. FAO, 2021. Fishery and Aquaculture Statistics. Global capture production 1950-2019
Size increment of jumbo flying squid Dosidicus gigas mature females in Peruvian (FishstatJ). In: FAO Fisheries Division [online]. Rome. Updated 2021. www.fao.org/
waters, 1989–2004. Prog. Oceanogr. 79, 308–312. https://doi.org/10.1016/j. fishery/statistics/software/fishstatj/en. Accessed 06/23/2022.
pocean.2008.10.003. Fisher, B., 2020. Controversy descends on Pacific sardine fishery over stock surveys.
Arkhipkin, A., Argüelles, J., Shcherbich, Z., Yamashiro, C., 2015b. Ambient temperature SeafoodSource News, April 23, 2020. https://www.seafoodsource.com/news/
influences adult size and life span in jumbo squid (Dosidicus gigas). Can. J. Fish. environment-sustainability/controversy-descends-on-pacific-sardine-fishery-over-
Aquat. Sci. 72, 400–409. https://doi.org/10.1139/cjfas-2014-0386. stock-surveys. Accessed 06/23/2022.
Arkhipkin, A.I., Rodhouse, P.G.K., Pierce, G.J., Sauer, W., Sakai, M., Allcock, L., Frawley, T., Briscoe, D., Daniel, P., Briitten, G.L., Crowder, L., Robinson, C.J., Gilly, W.,
Arguelles, J., Bower, J.R., Castillo, G., Ceriola, L., Chen, C.-S., Chen, X., Diaz- 2019a. Impacts of a shift to a warm-water regime in the Gulf of California on jumbo
Santana, M., Downey, N., González, A.F., Granados Amores, J., Green, C.P., squid (Dosidicus gigas). ICES J. Mar. Sci. 76 (7), 2413–2426. https://doi.org/
Guerra, A., Hendrickson, L.C., Ibáñez, C., Ito, K., Jereb, P., Kato, Y., Katugin, O.N., 10.1093/icesjms/fsz133.
Kawano, M., Kidokoro, H., Kulik, V.V., Laptikhovsky, V.V., Lipinski, M.R., Liu, B., Frawley, T.H., Crowder, L.B., Broad, K., 2019b. Heterogeneous perceptions of social-
Mariátegui, L., Marin, W., Medina, A., Miki, K., Miyahara, K., Moltschaniwskyj, N., ecological change among small-scale fishermen in the central Gulf of California:
Moustahfid, H., Nabhitabhata, J., Nanjo, N., Nigmatullin, C.M., Ohtani, T., Pecl, G., implications for adaptive response. Front. Mar. Sci. 6, 78. https://doi.org/10.3389/
Perez, J.A.A., Piatkowski, U., Saikliang, P., Salinas-Zavala, C.A., Steer, M., Tian, Y., fmars.2019.00078.
Ueta, Y., Vijai, D., Wakabayashi, T., Yamaguchi, T., Yamashiro, C., Yamashita, N., Galindo-Cortes, G., De Anda-Montañez, J.A., Arreguín-Sánchez, F., Salas, S., Balart, E.F.,
Zeidberg, L.D., 2015a. World squid fisheries. Rev. Fish. Sci. Aquacult. 23 (2), 2010. How do environmental factors affect the stock–recruitment relationship? The
92–252. case of the Pacific sardine (Sardinops sagax) of the northeastern Pacific Ocean. Fish.
Arreguín-Sánchez, F., del Monte-Luna, P., Albañez-Lucero, M.O., Zetina-Rejón, M.J., Res. 102, 173–183. https://doi.org/10.1016/j.fishres.2009.11.010.
Tripp-Quezada, A., Ruiz-Barreiro, T.M., Hernández-Padilla, J.C., 2021. How much García-Fernández, F., Gómez-Gutiérrez, J., Haskpiel-Segura, C., Ambriz-Arreola, I., De
biomass must remain in the sea after fishing to preserve ecosystem functioning? The Silva-Dávila, R., Martínez-López, A., Sánchez-Uvera, A., Hernández-Rivas, M.,
case of the sardine fishery in the Gulf of California, Mexico. In: Ortiz, M., Jordán, F. Robinson, C.J., This issue. Interannual response of euphausiid community
(Eds.), Marine Coastal Ecosystems Modelling and Conservation. Springer Nature, abundance during a prolonged anomalous warming period (June 2013–2019) in the
Switzerland, pp. 143–160. https://doi.org/10.1007/978-3-030-58211-17. Gulf of California. Prog. Oceanogr., Special volume Resilience of the Gulf of
Baumgartner, T.R., Christensen, N., 1985. Coupling of the Gulf of California to large- California to Climate Change.
scale interannual climatic variability. J. Mar. Res. 43 (4), 825–848. https://doi.org/ Gentemann, C.L., Fewings, M.R., García-Reyes, M., 2017. Satellite sea surface
10.1357/002224085788453967. temperatures along the West Coast of the United States during the 2014–2016
Bazzino, G., Salinas-Zavala, C., Markaida, U., 2007. Variability in the population northeast Pacific marine heat wave. Geophys. Res. Lett. 44, 312–319. https://doi.
structure of jumbo squid (Dosidicus gigas) in Santa Rosalia, central Gulf of California. org/10.1002/2016GL071039.
Cienc. Mar. 33, 173–186. https://doi.org/10.7773/cm.v33i2.1055. Gilly, W.F., Markaida, U., Baxter, C.H., Block, B.A., Boustany, A., Zeidberg, L.D.,
Bazzino, G., Gilly, W.F., Markaida, U., Salinas-Zavala, C.A., Ramos-Castillejos, J., 2010. Reisenbichler, K., Robison, B., Bazzino, G., Salinas, C., 2006. Vertical and horizontal
Horizontal migrations, vertical habitat utilization and diet of the jumbo squid migrations by the jumbo squid Dosidicus gigas revealed by electronic tagging. Mar.
(Dosidicus gigas) in the Pacific Ocean off Baja California Sur. Mexico. Prog. Oceanogr. Ecol. Prog. Ser. 324, 1–17. https://doi.org/10.3354/meps324001.
86, 59–71. https://doi.org/10.1016/j.pocean.2010.04.017. Gilly, W.F., Zeidberg, L.D., Booth, J.A.T., Stewart, J.S., Marshall, G., Abernathy, K.,
Beman, J.M., Arrigo, K.R., Matson, P.A., 2005. Agricultural runoff fuels large Bell, L., 2012. Locomotion and behavior of Humboldt squid, Dosidicus gigas, in
phytoplankton blooms in vulnerable areas of the ocean. Nature 434, 211–214. relation to natural hypoxia in the Gulf of California. Mexico. J. Exp. Biol. 215,
https://doi.org/10.1038/nature03370. 3175–3190. https://doi.org/10.1242/jeb.072538.
Bond, N.A., Cronin, M.F., Freeland, H., Mantua, N., 2015. Causes and impacts of the Gilly, W.F., Beman, J.M.B., Litvin, S.Y., Robison, B.H., 2013. Oceanographic and
2014 warm anomaly in the NE Pacific. Geophys. Res. Lett. 42, 3414–3420. https:// biological effects of shoaling of the oxygen minimum zone. Annual Rev. Mar. Sci. 5,
doi.org/10.1002/2015GL063306. 393–420. https://doi.org/10.1146/annurev-marine-120710-100849.
Brinton, E., Townsend, A.W., 1980. Euphausiids in the Gulf of California—the 1957 Gilly, W.F., Markaida, U., Robinson, C., Liu, B., 2020. Got Squid? Changes in the
cruises. Accessed 06/23/2022 Calif. Coop. Ocean. Fish. Invest. Rep. 21, 211–236. ecological state of the Gulf of California since the 1940 Steinbeck-Ricketts expedition
https://calcofi.org/publications/calcofi-reports/. with a focus on Humboldt squid. J. Southwest 62, 411–449. https://doi.org/
Brinton, E., Fleminger, A., Siegel-Causey, D., 1986. The temperate and tropical 10.1353/jsw.2020.0011.
planktonic biotas of the Gulf of California. Accessed 06/23/2022 Calif. Coop. Ocean. Giron-Nava, A., Ezcurra, E., Brias, A., Velarde, E., Deyle, E., Cisneros-Montemayor, A.M.,
Fish. Invest. Rep. 27, 228–266. https://calcofi.org/publications/calcofi-reports/. Munch, S.B., Sugihara, G., Aburto-Oropeza, A., 2021. Environmental variability and
Brownell, R.L., Reeves, R.R., Read, A.J., Smith, B., Thomas, P.O., Ralls, K.R., Amano, M., fishing effects on the Pacific sardine fisheries in the Gulf of California. Can. J. Fish.
Berggren, P., Chit, M.Y., Collins, T., Currey, R., Dolar, M.L.M., Genov, T., Hobbs, R. Aquat. Sci. 78, 623–630. https://doi.org/10.1139/cjfas-2020-0010.
C., Kreb, D., Marsch, H., Zhigang, M., Perrin, W.F., Phay, S., Rojas-Bracho, L., Gómez-Ocampo, E., Gaxiola-Castro, G., Durazo, R., Beier, E., 2017. Effects of the
Ryan, G.E., Shelden, K.E.W., Slooten, E., Taylor, B.L., Vidal, O., Ding, W., Whitty, T. 2013–2016 warm anomalies on the California Current phytoplankton. Deep Sea Res.
S., Yang, J.Y., 2019. Bycatch in gillnet fisheries threatens critically endangered small Part II 151, 64–76. https://doi.org/10.1016/j.dsr2.2017.01.005.
cetaceans and other aquatic megafauna. Endang. Species Res. 40, 285–296. https:// González-Silvera, A., Santamaría-del-Ángel, E., Camacho-Ibar, V., López-Calderón, J.,
doi.org/10.3354/esr00994. Santander-Cruz, J., Mercado-Santana, A., 2020. The effect of cold and warm
Cade, D.E., Benoit-Bird, K.J., 2015. Depths, migration rates and environmental anomalies on phytoplankton pigment composition in waters off the Northern Baja
associations of acoustic scattering layers in the Gulf of California. Deep Sea Res. Part California Peninsula (México): 2007–2016. J. Mar. Sci. Eng. 8, 533. https://doi.org/
I: Oceanogr. Res. Pap. 102, 78–89. https://doi.org/10.1016/j.dsr.2015.05.001. 10.3390/jmse8070533.
Capotondi, A., Wittenberg, A.T., Newman, M., Di Lorenzo, E., Yu, J.-Y., Braconnot, P., Gulland, F., Danil, K., Bolton, J., Ylitalo, G., Okrucky, R.S., Rebolledo, F., Alexander-
Cole, J., Dewitte, B., Giese, B., Guilyardi, E., Jin, F.-F., Karnauskas, K., Kirtman, B., Beloch, C., Brownell, R.l., Mesnick, S., Lefebvre, K., Smith, C.R., Thomas, P.O., Rojas-
Tong, L., Schneider, N., Xue, Y., Yeh, S.-W., 2015. Understanding ENSO diversity. Bracho, L., 2020. Vaquitas (Phocoena sinus) continue to die from bycatch not
Bull. Am. Meteorol. Soc. 96 (6), 921–938. https://doi.org/10.1175/BAMS-D-13- pollutants. Vet. Record 187 (7), e51.
00117.1. Hammann, M.G., Nevarez-Martínez, M.O., Green-Ruíz, Y.G., 1998. Spawning habitat of
Cavole, L.M., Demko, A.M., Diner, R.E., Giddings, A., Koester, I., Pagniello, C.M.L.S., the Pacific sardine (Sardinops sagax) in the Gulf of California: Egg and larval
Paulsen, M.-L., Ramirez-Valdez, A., Schwenck, S.M., Yen, N.K., Zill, M.E., Franks, P. distribution 1956-1957 and 1971-1991. Calif. Coop. Ocean. Fish. Invest. Rep. 39,
J.S., 2016. Biological impacts of the 2013–2015 warm-water anomaly in the 169–179. https://calcofi.org/publications/calcofi-reports/. Accessed 06/23/2022.
Northeast Pacific: Winners, losers, and the future. Oceanography 29 (2), 273–285. Hidalgo-González, R.M., Alvarez-Borrego, S., 2004. Total and new production in the Gulf
https://doi.org/10.5670/oceanog.2016.32. of California estimated from ocean color data from the satellite sensor SeaWIFS.
Chen, X., Li, J., Liu, B., Chen, Y., Li, G., Fang, Z., Tian, S., 2013. Age, growth and Deep Sea Res. Part I: Oceanog. Res. Papers 51, 739–752. https://doi.org/10.1016/j.
population structure of jumbo flying squid, Dosidicus gigas, off the Costa Rica Dome. dsr2.2004.05.006.
J. Mar. Biol. Ass. U.K. 93, 567–573. https://doi.org/10.1017/S0025315412000422. Hoving, H.-J., Gilly, W.F., Markaida, U., Benoit-Bird, K.J., West-Brown, Z., Daniel, P.,
Cimino, M.A., Jacox, M.G., Bograd, S.J., Brodie, S., Carroll, G., Hazen, E.L., Field, J.C., Parassenti, L., Liu, B., Campos, B., 2013. Extreme plasticity in life-history
Lavaniegos, B.E., Morales, M.M., Satterthwaite, E., Rykaczewski, R.R., 2021. strategy allows a migratory predator (Dosidicus gigas) to cope with a changing
Anomalous poleward advection facilitates episodic range expansions of pelagic red climate. Glob. Change Biol. 19, 2089–2103. https://doi.org/10.1111/gcb.12198.
crabs in the eastern North Pacific. Limnol. Oceanogr. 66 (8), 3176–3189. https://doi. Howard, E.M., Penn, J.L., Frenzel, H., Seibel, B.A., Bianchi, D., Renault, L., Kessouri, F.,
org/10.1002/lno.11870. Sotula, M.A., McWilliam, J.C., Deutsch, C., 2020. Climate-driven aerobic habitat loss

19
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

in the California Current System. Science 6 (20), eaay3188. https://doi.org/ Páez-Osuna, F., Álvarez-Borrgo, S., Ruiz-Fernández, A.C., GarcÍa-Hernández, J., Jara-
10.1126/sciadv.aay3188. Marini, M.E., Bergés-Tiznado, M.E., Piñón-Gimate, A., Alonso-Rodríguez, R., Soto-
Izquierdo-Peña, V., Lluch-Cota, S.E., Chavez, F.P., Lluch-Cota, D.B., Morales- Jiménez, M.F., Frías-Espericueta, M.G., Ruelas-Inzunza, J.R., Green-Ruiz, C.R.,
Bojórquez, E.M., Ponce-Díaz, G., 2020. Is there a future in the sustainability Osuna-Martínez, C.C., Sanchez-Cabeza, J.-A., 2017. Environmental status of the Gulf
certification of sardine and anchovy fisheries? Fisheries 45, 554–560. https://doi. of California: A pollution review. Earth-Sci. Rev. 166, 181–205. https://doi.org/
org/10.1002/fsh.10461. 10.1016/j.earscirev.2017.01.014.
Jaquet, N., Gendron, D., 2002. Distribution and relative abundance of sperm whales in Petatán-Ramírez, D., Ojeda-Ruiz, M.A., Sánchez-Velasco, L., Rivas, D., Reyes-Bonilla, H.,
relation to key environmental features, squid landings and the distribution of other Cruz-Piñón, G., Morzaria-Luna, H.N., Cisneros-Montemayor, A.M., Cheung, W.,
cetacean species in the Gulf of California. Mexico. Mar. Biol. 141, 591–601. https:// Salvadeo, C., 2019. Potential changes in the distribution of suitable habitat for
doi.org/10.1007/s00227-002-0839-0. Pacific sardine (Sardinops sagax) under climate change scenarios. Deep-Sea Res. Part
Keyl, F., Wolff, M., Arguelles, J., Mariategui, L., Tafur, R., Yamashiro, C., 2008. II 169–170, 104632. https://doi.org/10.1016/j.dsr2.2019.07.020.
A hypothesis on range expansion and spatio temporal shifts in size at maturity of Portela, E., Beier, E., Barton, E.D., Castro, R., Godínez, V., Palacios-Hernández, E.,
jumbo squid (Dosidicus gigas) in the Eastern Pacific Ocean. Accessed 06/23/2022 Fielder, P.C., Sanchez-Velasco, L., Trasviña, A., 2016. Water masses and circulation
Calif. Coop. Ocean. Fish. Invest. Rep. 49, 119–128. https://calcofi.org/publications/ in the tropical Pacific off central Mexico and surrounding areas. J. Phys. Oceanogr.
calcofi-reports/. 46 (10), 3069–3081. https://doi.org/10.1175/JPO-D-16-0068.1.
Keyl, F., Argüelles, J., Tafur, R., 2011. Interannual variability in size structure, age, and Portner, E., Benoit-Bird, K.J., Hazen, E.L., Waluk, C.M., Robinson, C.J., Gómez-Gutiérrez,
growth of jumbo squid (Dosidicus gigas) assessed by modal progression analysis. ICES J., Gilly, W., 2022. Depression and recovery of pelagic acoustic backscattering
J. Mar. Sci. 68, 507–518. https://doi.org/10.1093/icesjms/fsq167. following El Niño events in the Gulf of California. Prog. Oceanogr., special volume
Lavín, M.F., Marinone, S.G., 2003. An overview of the physical oceanography of the Gulf Resilience of the Gulf of California to Climate Change. 2006, 102823. https://doi.
of California. In: Velasco, O.U., Sheinbaum, J., Ochoa, J. (Eds.), Nonlinear Processes org/10.1016/j.pocean.2022.102823.
in Geophysical Fluid Dynamics. Springer, New York, pp. 173–204. Portner, E.J., Markaida, U., Robinson, C.J., Gilly, W.F., 2019. Trophic ecology of
Lilly, L.E., Ohman, M.D., 2018. CCE IV: El Niño-related zooplankton variability in the Humboldt squid, Dosidicus gigas, in conjunction with body size and climatic
southern California Current System. Deep-Sea Res Part I 40, 36–51. https://doi.org/ variability in the Gulf of California. Mexico. Limnol. Oceanogr. 65, 732–748.
10.1016/j.dsr.2018.07.015. https://doi.org/10.1002/lno.11343.
Lilly, L.E., Ohman, M.D., 2021. Euphausiid spatial displacements and habitat shifts in the Radovich, J., 1982. Collapse of the California sardine fishery. Accessed 6/23/22 Calif.
southern California Current System in response to El Niño variability. Prog. Coop. Ocean. Fish. Invest. Rep. 23, 56–78. https://calcofi.org/publications/calcofi
Oceanogr. 193, 102544 https://doi.org/10.1016/j.pocean.2021.102544. -reports/.
Lipiínski, M.R., Underhill, L.G., 1995. Sexual maturation in squid: quantum or Ramírez-Rodríguez, M., Almendárez-Hernández, L.C., 2013. Subsidies in the jumbo squid
continuum? South African. J. Mar. Sci. 15, 207–223. https://doi.org/10.2989/ fishery in the Gulf of California. Mexico. Mar. Policy 40, 117–123. https://doi.org/
02577619509504844. 10.1016/j.marpol.2013.01.005.
Lluch-Cota, S.E., Parés-Sierra, A., Magaña-Rueda, V.O., Arreguín-Sánchez, F., Ricketts, E.F., 1947. Science studies the sardine. Monterey Peninsula Herald, 12 Annual
Bazzino, G., Herrera-Cervantes, H., Lluch-Belda, D., 2010. Changing climate in the Sardine Edition, March 7, p. 1,3. http://www.datadeluge.com/2009/06/science-
Gulf of California. Prog. Oceanogr. 87 (1–4), 114–126. https://doi.org/10.1016/j. studies-sardine-mysterious_27.html. Accessed 06/23/2022.
pocean.2010.09.007. Ricketts, E.F., 1948. Investigator blames industry, nature for shortage. Monterey
Markaida, U., 2006. Population structure and reproductive biology of jumbo squid Peninsula Herald, 13th Annual Sardine Edition, April 2, 1948, in Rodgers KA (ed),
Dosidicus gigas from the Gulf of California after the 1997–1998 El Niño event. Fish. Breaking through. Essays, Journals and Travelogues of Edward F. Ricketts, Univ. CA
Res. 79, 28–37. https://doi.org/10.1016/j.fishres.2006.02.009. Press, Berk5eley, pp. 325–330.
Markaida, U., Rosenthal, J.J.C., Gilly, W.F., 2005. Tagging studies of the Humboldt Robinson, C.J., Gómez-Gutiérrez, J., de León, D.A.S., 2013. Jumbo squid (Dosidicus gigas)
squid, Dosidicus gigas, in the Gulf of California. Accessed 06/23/2022 Mexico. Fish. landings in the Gulf of California related to remotely sensed SST and concentrations
Bull. 103, 219–226. https://www.semanticscholar.org/paper/Tagging-studies-on- of chlorophyll a (1998–2012). Fish. Res. 137, 97–103. https://doi.org/10.1016/j.
the-jumbo-squid-(Dosidicus-in-of-Markaida-Rosenthal/463e4f52f2e21cfe916b2ce84 fishres.2012.09.006.
f3be4b49e94ddf5. Robinson, C.J., Gomez-Gutierrez, J., Markaida, U., Gilly, W.F., 2016. Prolonged decline
Martínez-Porchas, M., Hernández-Rodríguez, M., Bückle-Ramírez, L.F., 2009. Thermal of jumbo squid (Dosidicus gigas) landings in the Gulf of California is associated with
behavior of the Pacific sardine (Sardinops sagax) acclimated to different thermal chronically low wind stress and decreased chlorophyll a after El Niño 2009–2010.
cycles. J. Thermal Biol. 34, 372–376. https://doi.org/10.1016/j. Fish. Res. 173, 128–138. https://doi.org/10.1016/j.fishres.2015.08.014.
jtherbio.2009.07.002. Rodhouse, P.K., Yamashiro, C., Arguelles, J., 2016. Jumbo squid in the eastern Pacific
Martínez-Porchas, M., Hernández-Rodríguez, M., Herzka, S.Z., Gutierrez-Millán, L.E., Ocean: A quarter century of challenges and change. Fish. Res. 173, 109–112. https://
2011. Evaluation of the physiological status of the Pacific sardine, Sardinops sagax doi.org/10.1016/j.fishres.2015.11.001.
caeruleus, acclimated to different thermal regimes based on selected blood Rojas-Bracho, L., Brusca, R.C., Álvares-Borrego, S., Brownell, R.L., Camacho-Ibar, V.,
parameters. Environ. Biol. Fish. 91, 39–49. https://doi.org/10.1007/s10641-010- Ceballos, G.C., De La Cueva, H., García-Hernández, J., Hastings, P.A., Cárdenas-
9757-z. Hinojosa, G., Jaramillo-Legorreta, A.M., Medllín, R., Mesnick, S.L., Nieto-García, E.,
Martínez-Soler, E., Gómez-Gutiérrez, J., de Silva-Dávila, R., González-Rodríguez, E., Urbán, J., Velarde, E., Vidal, O., Findley, L.T., Taylor, B.L., 2019. Unsubstantiated
Aburto-Oropeza, O., Dolan, J., 2021. Cephalopod paralarval species richness, claims can lead to tragic conservation outcome. Bioscience 69, 12–14. https://doi.
abundance, and size structure during the 2014–2017 anomalous warm period in the org/10.1093/biosci/biy138.
southern Gulf of California. J. Plankton Res. 43 (2), 224–243. Rosa, R., Seibel, B.A., 2008. Synergistic effects of climate-related variables suggest future
McFarlane, G.A., Macdougal, L., Scheigert, J., Hrabok, C., 2005. Distribution and biology physiological impairment in a top oceanic predator. Proc. Natl. Acad. Sci. USA 105,
of Pacific sardines (Sardinops sagax) off British Columbia, Canada. Calif. Coop. 20775–20780. https://doi.org/10.1073/pnas.0806886105.
Ocean. Fish. Invest. Rep. 46, 144–160. https://calcofi.org/publications/calcofi- Rosa, R., Seibel, B.A., 2010. Metabolic physiology of the Humboldt squid, Dosidicus gigas:
reports/. Accessed 06/23/2022. implications for vertical migration in a pronounced oxygen minimum zone. Prog.
Moore, K., 2021. Oceana sues NMFS over California sardine management. National Oceanogr. 86, 72–80. https://doi.org/10.1016/j.pocean.2010.04.004.
Fisherman. July 20, 2021. https://www.nationalfisherman.com/west-coast-pacific/ Rosas-Luis, R., Salinas-Zavala, C.A., Koch, V., Del Monte Luna, P., Morales-Zárate, M.V.,
oceana-sues-nmfs-over-california-sardine-management. Accessed 06/23/2022. 2008. Importance of jumbo squid Dosidicus gigas (Orbigny, 1835) in the pelagic
Morales-Bojórquez, E., Hernández-Herrera, A., Nevárez-Martínez, M.O., Díaz-Uribe, J.G., ecosystem of the central Gulf of California. Ecol. Model. 218 (1–2), 149–161.
2012. Population size of the jumbo squid Dosidicus gigas in the central Gulf of https://doi.org/10.1016/j.ecolmodel.2008.06.036.
California, Mexico, based on mark-recapture data. Aquat. Biol 15, 27–34. https:// Rubio-Rodríguez, U., Villalobos, H., Nevárez-Martínez, M.O., 2018. Acoustic
doi.org/10.3354/ab00406. observations of the vertical distribution and latitudinal range of small pelagic fish
Morrow, R.M., Ohman, M.D., Goericke, R., Kelly, T.B., Stephens, B.M., Stukel, M.R., schools in the Midriff Islands Region, Gulf of California. Mexico. Lat. Am. J. Aquat.
2018. CCE V: Primary production, mesozooplankton grazing, and the biological Res. 46 (5), 989–1000. https://doi.org/10.3856/vol46-issue5-fulltext-12.
pump in the California Current Ecosystem: Variability and response to El Niño. Deep- Rudnick, D.L., Zaba, K.D., Todd, R.E., Davis, R.E., 2017. A climatology of the California
Sea Res Part I 40, 52–62. https://doi.org/10.1016/j.dsr.2018.07.012. Current System from a network of underwater gliders. Prog. Oceanog. 154, 64–106.
Navarro-Olache, L.F., Lavín, M.F., Alvarez-Sánchez, L.G., Zirino, A., 2004. Internal https://doi.org/10.1016/j.pocean.2017.03.002.
structure of SST features in the central Gulf of California. Deep Sea Research Part II: Ruiz-Cooley, R.I., Engelhaupt, D.T., Ortega-Ortiz, J.G., 2012. Contrasting C and N
Topical Studies in Oceanography 51, 673–687. https://doi.org/10.1016/j. isotope ratios from sperm whale skin and squid between the Gulf of Mexico and Gulf
dsr2.2004.05.014. of California: effect of habitat. Mar. Biol. 159 (1), 151–164. https://doi.org/
Nevárez-Martínez, M.O., Lluch-Belda, D., Cisneros-Mata, M.A., Santos-Molina, J.P., 10.1007/s00227-011-1795-3.
Martínez-Zavala, M.A., Lluch-Cota, S.E., 2001. Distribution and abundance of the Sala, E., Aburto-Oropeza, O., Reza, M., Paredes, L.-L., L.,, 2004. Fishing down coastal
Pacific sardine (Sardinops sagax) in the Gulf of California and their relation with the food webs in the Gulf of California. Fisheries 29, 19–25. https://doi.org/10.1577/
environment. Prog. Oceanogr. 49, 565–580. https://doi.org/10.1016/S0079-6611 1548-8446(2004)29[19:FDCFWI]2.0.CO;2.
(01)00041-6. Schlitzer, R., 2021. Ocean Data View, https://odv.awi.de. Accessed 06/23/2022.
NOAA, 2015. El Niño/Southern Oscillation (ENSO) diagnostic discussion. Climate Seibel, B.A., 2013. The jumbo squid, Dosidicus gigas (Ommastrephidae), living in oxygen
Prediction Center, National Weather Service, March 5, 2005. https://www.cpc.ncep. minimum zones II: Blood–oxygen binding. Deep Sea Res. Part II: Top. Stud.
noaa.gov/products/analysis_monitoring/enso_disc_mar2015/ensodisc.html. Oceanogr. 95, 139–144. https://doi.org/10.1016/j.dsr2.2012.10.003.
Accessed 06/23/2022. Stewart, J., Field, J.C., Markaida, U., Gilly, W.F., 2013. Behavioral ecology of jumbo
Ohman, M.D., 2018. Introduction to collection of papers on the response of the southern squid (Dosidicus gigas) in relation to oxygen minimum zones. Deep Sea Res. II 95,
California Current ecosystem to the warm anomaly and El Niño, 2014–16. Deep-Sea 197–208. https://doi.org/10.1016/j.dsr2.2012.06.005.
Res. Part I 140, 1–3. https://doi.org/10.1016/j.dsr.2018.08.011.

20
W. Gilly et al. Progress in Oceanography 206 (2022) 102857

Stramma, L., Johnson, G.C., Sprintall, J., Mohrholz, V., 2008. Expanding oxygen- Unesco, 2021. Concerns over the Vaquita in the Islands and Protected Areas of the Gulf of
minimum zones in the tropical oceans. Science 320, 655–658. https://doi.org/ California (Mexico). Accessed 6/22/2022. https://whc.unesco.org/en/news/2310/.
10.1126/science.1153847. UNESCO, 2005. Decision: 29 COM 8B.9. Nominations of Natural Properties to the World
Stramma, L., Schmidtko, S., Levin, L.A., Johnson, G.C., 2010. Ocean oxygen minima Heritage List (Islands and Protected Areas of the Gulf of California). https://whc.
expansions and their biological impacts. Deep-Sea Res. I 57, 587–595. https://doi. unesco.org/en/decisions/472. Accessed 6/22/2022.
org/10.1016/j.dsr.2010.01.005. Vaquer-Sunyer, R., Duarte, C.M., 2008. Thresholds of hypoxia for marine biodiversity.
Tafur, R., Keyl, F., Argüelles, J., 2010. Reproductive biology of jumbo squid Dosidicus Proc. Natl. Acad. Sci. USA 105, 15452–15457. https://doi.org/10.1073/
gigas in relation to environmental variability of the northern Humboldt Current pnas.0803833105.
System. Mar. Ecol. Prog. Ser. 400, 127–141. https://doi.org/10.3354/meps08386. Velarde, E., Ezcurra, E., Horn, M.H., Patton, R.T., 2015. Warm oceanographic anomalies
Unesco, 2019. The Islands and Protected Areas of the Gulf of California (Mexico) and fishing pressure drive seabird nesting north. Sci. Adv. 1, e1400210 https://doi.
inscribed on the List of World Heritage in Danger. Accessed 6/22/2022. https://whc. org/10.1126/sciadv.1400210.
unesco.org/en/news/1999. Velarde, E., Anderson, D.W., Ezcurra, E., 2019. Seabird clues to ecosystem health.
Science 365 (6449), 116–117. https://doi.org/10.1126/science.aaw9999.

21