There are no data available on the global population size of the Great Hammerhead. Genetic studies support two subpopulations of Great Hammerhead, the Atlantic and Indo-Pacific (Testerman 2014). No genetic structure was found within the Atlantic or within the Indo-Pacific between eastern and western Australia (Testerman 2014).

Population trend data are available from three sources: (1) stock assessment in the Northwest Atlantic and Gulf of Mexico (Jiao et al. 2011), and from standardized catch-per-unit-effort (CPUE) in: (2) Northwest Atlantic and Gulf of Mexico (J. Carlson unpubl. data) and, (3) the Indian Ocean (Dudley and Simpfendorfer 2006). The trend data from each source were analyzed over three generation lengths using a Bayesian state-space framework (a modification of Winker et al. 2018). This analysis yields an annual rate of change, a median change over three generation lengths, and the probability of the most likely IUCN Red List category percent change over three generations (see the Supplementary Information).

First, the Northwest Atlantic stock assessment indicated that the stock was overfished from the mid-1980s with overfishing occurring periodically from 1983–1997 and the risk of overfishing low after 2001 (Jiao et al. 2011). A previous stock assessment also found the stock overfished but experiencing overfishing after 2001 (Hayes 2008); however, both stock assessments report a high degree of uncertainty mostly due to poor catch data (Miller et al. 2014). The trend analysis of the Northwest Atlantic modeled abundance for 1981–2005 (25 years) revealed annual rates of reduction of 0.4%, consistent with an estimated median reduction of 29.1% over three generation lengths (74.4 years), with the highest probability of <20% reduction over three generation lengths.

Second, more recent data (1994–2017) are available from the Northwest Atlantic and Gulf of Mexico comprising two of the time-series underlying the Jiao et al. (2011) stock assessment (J. Carlson unpubl. data). Both time-series indicate this population has begun to increase soon after the implementation of management after 2005 (NMFS 2006). The annual fisheries-independent bottom longline surveys (Grace and Henwood 1998) were conducted throughout the northern Gulf of Mexico and Southeast Atlantic by the National Marine Fisheries Service, Southeast Fisheries Science Center, Mississippi Laboratories; called the NMFS Mississippi bottom longline shark survey (NMFS-LL-SE; Ingram et al. 2005). These data (red in the Supplementary Information) exhibit an increase in CPUE from 2005 onwards. Additionally, the commercial shark bottom longline fishery is active in the U.S. Atlantic Ocean from around North Carolina to Florida and throughout the eastern Gulf of Mexico (BLLOP) (Morgan et al. 2009). These data (green in the Supplementary Information) exhibited an increasing but variable trend. Both time-series taken together for 1994–2017 (24 years) demonstrated a median increase and the highest probability of an increase over three generation lengths (74.4 years).

Third, the analysis of population abundance from the Indian Ocean Natal Shark’s Board bather protection netting program for 1978–2003 (26 years) (Dudley and Simpfendorfer 2006) yielded annual rates of reduction of 6.5% consistent with a median reduction of 99.3% over three generation lengths (71.1 years), with the highest probability of >80% reduction over three generation lengths

In addition to these population trend analyses, we considered a recent analysis of fishery-dependent longline CPUE from the South Atlantic. These time-series are complex, reflecting access by different fleets and nationalities over time, but the best estimates of trends come from phase B spanning 1998 to 2008 during which 20 fleets (170 vessels) fished for tunas, swordfishes, and sharks. During this 10-year period there was a 61.7% decline in CPUE of hammerheads (Sphyrna spp.) (Table S11, Barreto et al. 2016), however, because the data were not considered robust due to the low catch rates and large confidence intervals (R. Barreto unpubl. data), they were not analyzed over three generations.

Steep declines of hammerheads (Sphyrna spp.) have occurred in the Mediterranean Sea; Ferretti et al. (2008) compiled nine time-series of abundance indices from commercial and recreational fishery landings, scientific surveys and sighting records, to reconstruct long-term population trends of large sharks in the northwest Mediterranean Sea. Of the taxa for which there were enough data to investigate, hammerhead sharks declined the fastest; disappearing from coastal waters after 1963 followed by declines (across all fishing sectors) in pelagic waters by the early 1980s. Meta-analysis showed an average instantaneous rate of decline in abundance of -0.17 (time range 178 years) and biomass of -0.36 (time range 107 years), which equates to an estimated decline of 99.99% in abundance and biomass since the early 19th century (Ferretti et al. 2008). Walker et al. (2005) also report that hammerhead species have virtually disappeared from the central-southern Mediterranean Sea since 1986.

Population trend data are lacking from the North Pacific and are limited from the Western Central Pacific, where hammerhead catches are for the species-complex, with no indication of the proportion of Great Hammerheads. The CPUE (1995–2015) was highly variable and as Great Hammerhead are captured infrequently, little inference was possible from the temporal trend (Rice et al. 2015). Fishing pressure has significantly increased in recent decades across the distribution range of the Great Hammerhead in the tropical Indian Ocean and West Africa (Cooke 1997, Bruckner et al. 2011, Belhabib et al. 2012, ANSD 2016, Jabado et al. 2017), and has likely caused declines in these regions.

The Great Hammerhead was estimated to have undergone steep declines in the Indian Ocean with the Northwest Atlantic and Gulf of Mexico showing historic declines but some signs of stabilization and possible recovery in response to management. The Atlantic subpopulation is inferred to have undergone a >50% reduction over three generation lengths (74.4 years); although there is possible recovery in the Northwest Atlantic, there is also a high degree of uncertainty in the data and high levels of exploitation. The Indo-Pacific subpopulation is inferred to have undergone a >80% reduction over three generation lengths (71.1 years). To estimate a global population trend, the estimated three generation population trends for the North Atlantic and Indian Ocean regions were weighted according to the relative size of each region; the two sources of Atlantic data were used to generate two global trends. This resulted in a median reduction of 50.9–62.4%, with the highest probability of >80% reduction in both trends over three generation lengths (71.1–74.4 years). However, the area of some time-series is limited relative to the region and may not accurately represent the trend in Great Hammerhead across the entire region. There is uncertainty in some of the catch data and levels of exploitation. Exploitation levels may be similar to those of those of the Scalloped Hammerhead, though this is not definitively known, and the regions with no trend data are likely areas where the species is under intensive fishing pressure, such as the tropical Indian Ocean and West Africa. Hence, a precautionary approach was taken and expert judgement elicitation inferred a global population reduction of >80% over three generation lengths (71.1–74.4 years). Therefore, the Great Hammerhead is assessed as Critically Endangered A2bd.