There are no data available on the population size of the Scalloped Hammerhead. Genetic data reveal that the global population structure varies between males and females. Only males move across ocean basins, while females only move regionally and not between discontinuous continental coastlines (Duncan et al. 2006, Daly-Engel et al. 2012).

Four sets of population trend data are available: (1) a 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 and W.B. Driggers unpubl. data); (3) the South Pacific (Simpfendorfer et al. 2010); and, (4) the Indian Ocean (Dudley and Simpfendorfer 2006). For the South Pacific, there is bather protection mesh and drumline CPUE (Noriega et al. 2011), yet these data spanned only one decade and were not considered in the global analysis. The above trend data were analysed 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 Red List Category percent change over three generations (see Supplementary Material).

First, the Northwest Atlantic and Gulf of Mexico stock assessment indicated that the stock was overfished (B<Bmsy) from the early 1980s with overfishing (F>Fmsy) occurring periodically from 1983–2005 (Jiao et al. 2011). The steepest declines occurred prior to 1995, thereafter the abundance index remained stable until the end of the time series in 2005. The trend analysis of the modelled population abundance for 1981–2005 (25 years) yielded annual rates of reduction of 7.5% consistent with an estimated median reduction of 99.6% over three generation lengths (72.3 years), with the highest probability of >80% 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 and W.B. Driggers unpubl. data). Both time-series indicate this population has begun to increase soon after the implementation of management. The annual fisheries-independent bottom longline surveys (Grace and Henwood 1998) were conducted throughout the northern Gulf of Mexico and Northwest Atlantic by the National Marine Fisheries Service (NMFS), Southeast Fisheries Science Center, Mississippi Laboratories; called the NMFS Mississippi bottom longline shark survey (NMFS-LL-SE, in number of sharks per 10,000 hook hours; Ingram et al. 2005). These data (red in the Supplementary Information) exhibit a steep increase in CPUE from 2010 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 a slight, but variable increasing 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 (72.3 years).

Third, the analysis of the South Pacific Queensland Shark Control Program’s bather protection netting program population abundance for 1964–2004 (41 years) yielded annual rates of reduction of 8.4% consistent with an estimated median reduction of 99.8% over three generation lengths (72.3 years), with the highest probability of >80% reduction over three generation lengths.

Fourth, the analysis of the Indian Ocean Natal Sharks Board bather protection netting program population abundance for 1978–2003 (26 years) yielded annual rates of reduction of 4.0% consistent with an estimated median reduction of 93.4% over three generation lengths (72.3 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 (100 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 analysed over three generations.

Across the regions, the Scalloped Hammerhead was estimated to have undergone steep declines in all oceans, with some signs of stabilization and possible recovery in response to management only in the Northwest Atlantic and Gulf of Mexico. To estimate a global population trend, the estimated three generation population trends for each region 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 76.9–97.3%, with the highest probability of >80% reduction over three generation lengths (72.3 years), and therefore the Scalloped Hammerhead is assessed as Critically Endangered A2bd.