Dataset: Acropora hyacinthus energetics and reproduction

Methodology and analytical procedures documented in Leinbach et al., 2021, but are included here as well:
In tagged coral colonies that were resistant to or recovered from bleaching, the energetic condition of the host was assessed (Nresistant = 12, Nrecovered = 20). One small branch (~ 2–4 centimeters in length) was sampled from each colony and airbrushed in filtered seawater to remove coral tissue and algal cells (blastate) from the skeleton. The blastate was homogenized and 200 μL was collected and preserved in Z-fix (10% zinc formalin) for algal symbiont counts. The remaining blastate was centrifuged at 2000×g for 2 minutes to separate the host tissue from endosymbiont cells. Host tissue slurry was preserved at − 20 °C until further processing. Microalgal endosymbiont density was quantified using a hemocytometer (Hausser Scientific, Horsham, PA) under an Olympus BH-2 microscope. Total host protein content was quantified using a Bradford assay with bovine-serum albumin (BSA) as a standard (Pierce Coomassie Plus Assay Kit, Thermo Fisher Scientific). Total host carbohydrate content was quantified using a modified phenol–sulfuric acid method.  All physiological metrics were standardized to coral skeleton surface area following the paraffin wax-dipping technique.

Small fragments from tagged colonies were sampled by hand via SCUBA (Nresistant = 26, Nrecovered = 21) in October 2019. For each colony, the selected fragment was sampled 5–10 centimeters (cm) from the colony edge, and branch tips and colony edges were avoided. Samples were immediately preserved in Z-fix for 24 hours and then stored in 100% ethanol until histological processing. Samples were decalcified with a 1% EDTA decalcifier solution for 48–72 h and stored in 70% ethanol until processing on a Leica ASP6025 tissue processer. Paraffinized tissue was embedded in wax blocks (Leica EG1150H embedding machine) and then allowed to cool in a freezer 24 hours prior to sectioning. Blocks were serially sectioned at 5 micrometer (μm) thickness on a Leica RM2125RTS microtome every 300 micrometer (μm), which corresponds to the average oocyte diameter. Sections were arranged on microscope slides and stained using a modified Heidenhain’s aniline blue stain on a Leica ST5020 multistainer.

Histological sections were analyzed for measurements of reproductive effort: (1) presence/absence of male and female gametes, (2) diameter of oocytes, and (3) relative fecundity, detailed below. Gametes (oocytes and spermatocytes) were staged from I-V following the classification of Szmant-Froelich et al. Slides were examined using an Olympus BX41 microscope with an Olympus SC180 camera attachment. Measurements were made using ImageJ. Oocyte diameter was determined by averaging the longest and shortest axis of each oocyte. A total of 25 oocytes were measured from each colony. In fragments containing fewer than 25 oocytes, the maximum number of oocytes observed was used (Supplementary Table S1, Leinbach et al., 2021). Only oocytes with a visible nucleus were measured to ensure no oocytes were counted more than once and that the maximum diameter was measured.

Due to the small size of the fragments and polyps, as a proxy for fecundity, three polyps were randomly selected on the middle slide from each individual. When there was an even number of slides, the first of the two middle slides were used. Because only one slide from each individual was examined, there was no risk of double-counting oocytes, so the number of both nucleated and non-nucleated oocytes was counted in each of the randomly selected polyps. These counts were averaged to produce the average number of oocytes per polyp for each individual as a measure of relative fecundity. It should be noted that this relative estimate is lower than true fecundity.