Dataset: Larval Atlantic Bluefin Tuna prey from gut analysis, collected on NOAA Ship R/V Nancy Foster cruises NF1704 and NF1802 in the Gulf of Mexico, May 2017 and May 2018.

Larval sampling:
Larval Atlantic Bluefin Tuna (ABT) were collected during the peak spawning period in the GoM during May of 2017 (NF1704) and 2018 (NF1802). Samples were taken with a dual 90 cm diameter bongo frame (bongo-90) equipped with 500-µm synthetic nylon mesh nets by oblique hauls in the upper 25 m of the water column and towed at an average over-ground speed of 2.22 knots for 10 minutes (Habtes et al., 2014; Laiz-Carrion et al.,2015). Filtered volumes were calculated with a mechanical flowmeter (2030R, General Oceanics Inc.) placed in the center of the net mouths. Larval patches were located by transect net sampling and real-time sample sorting (every ~10 nautical miles) across favorable habitat (Domingues et al., 2016) with a higher probability of containing ABT larvae. Once larvae were detected, the core of the patch was located and marked with a drogued satellite-tracked drifter equipped with a strobe light, beginning a 3–4 day experimental cycle during which the ABT larvae were sampled close to the drifter every 3 hours. Three cycle experiments (C1, C2, and C3) were conducted on the NF1704 cruise, and two cycles (C4 and C5) were done on NF1802—ABT larvae were only abundant during C1 and C5.

Upon collection, left bongo-90 samples were immediately fixed in 95% EtOH and the right bongo-90 samples were immediately sorted for ABT larvae. Contents were concentrated on a sieve using 50-µm filtered cold seawater and placed in a large petri dish in a Styrofoam box with ice while sorting. Using a dissecting microscope (MZ12, Leica Microsystems), ABT larvae were removed from small portions of the wet sample that could be sorted within 10 minutes and placed in a small petri dish in the cooler. The remaining zooplankton in the sample were immediately fixed with 95% EtOH. The collected zooplankton volume was generally small (<250 mL during the day), but some night samples exceeded 1 L. For larger samples, sorting was halted within 1 hour and fixed to prevent sample degradation; sorting was resumed back in the lab. Fixative was refreshed after 24 hours to prevent sample degradation from dilution.

Gut analysis:
The alimentary canal from pharynx to anus was dissected using tweezers and scalpel by carefully tracing and cutting along the dorsal margin of the gut from anus to operculum, just deep enough to allow access to the organs. Another deeper cut was made under the operculum and gills to separate the pharynx from the head, and tweezers were used to open the lateral muscular membrane and pick out the internal organs. The digestive tract was carefully opened from anus to stomach using sharpened tips of insect pins, and contents were isolated and imaged. We recorded the location of each prey item in three sections of the alimentary canal: foregut (from pharynx to stomach), midgut (anterior intestine, a large ventral pouch), and hindgut (posterior intestine, a narrow tube connecting dorsally over the right side of the stomach to the anus).