Location:
patch reef 13 in Kāne'ohe Bay, O'ahu, Hawai'i (21.4509, -157.7954).

This dataset provides bleaching score data and mortality.  The following section also describes the closely related dataset "Benthic cover" https://www.bco-dmo.org/dataset/897403 which provides the community composition point counts and images.

Coral colony selection:

All colonies followed in this study were first categorized as bleaching-susceptible (severely bleached) or bleaching-resistant (fully pigmented) based on their bleaching phenotype during the peak of the 2015 heatwave and coral bleaching event (Matsuda et al., 2020). For this study, ten pairs of adjacent conspecific colonies of M. capitata and P. compressa with contrasting bleaching susceptibilities (N=10 colonies per species per phenotype) were selected. Adjacent pairs of bleaching-resistant and bleaching-susceptible colonies of the same species were selected in order to minimize the influence of microenvironment on the bleaching response. Individual colonies were monitored for bleaching (color/pigmentation) and partial mortality from 2015–2017 (Matsuda et al., 2020) and 2019–2023 (this study), and sampled for physiological assessments from 2019–2023 (Brown et al., 2023 Tables S3 and S4). An additional pair of M. capitata and two pairs of P. compressa were added to the time series in 2022 to supplement our observations after three pairs could no longer be located; all of these colonies had been assessed for bleaching, mortality and recovery from 2015–2017.

Nearly all of the M. capitata colonies used in this study (20 of the 22 colonies) were identified as unique genotypes in an earlier study (Drury, 2022). In general, clonality in M. capitata in Kāne‘ohe Bay is very low (Caruso, et al., 2022) with a bay-wide genet-ramen ratio of 0.917. Caruso et al (2022) included two sites at the same reef investigated in this study (Patch Reef 13), identifying a genet-ramen (G:R) ratio of 0.95 (i.e., 21 genotypes in 22 colonies sampled). For P. compressa, the bay-wide genet-ramen ratio is approximately 0.875, but clonality is rare in low wave energy (inner bay) environments (Locatelli, et al., 2019). Similarly, P. compressa from sheltered South Bay sites have a genet-ramen ratio of 0.96 (Hunter, 1993). The likelihood of there being more than three clones in P. compressa is very low (0.92 [average of two papers G:R] * 24 colonies = 22 genotypes), especially considering the physiological variation observed. Overall, these studies indicate infrequent asexual reproduction at the study site for either species. 

Coral bleaching and partial mortality assessments:

Colony-level bleaching severity was determined from photographs of each colony following the methodology of (Innis et al., 2021), in which colonies were scored as: (1) no signs of paling (0%), (2) mild paling (>20%), (3) moderate paling (20–50%), (4) mostly bleached (50–80%), and (5) fully bleached (80–100%). Cumulative colony-level partial mortality was also determined from these same photographs as described in  (Matsuda et al., 2020). Observations occurred during peak and off-peak seasonal temperatures in most years. Benthic community composition was determined at the same time as colony-level observations following the same methods as in  (Matsuda et al., 2020; Innis et al., 2021). Specifically, benthic photoquadrats (0.33 m2), were imaged at 2 m intervals along a 40 m transect tape laid parallel to the reef crest at 1 m and 3 m depths (n = 1–2 per depth) at PR13. Benthic community composition was determined from each image via CoralNet using 50 randomly allocated points per photograph (Beijbom, 2015). Bleaching severity of each coral point was scored as: (1) pigmented (no signs of bleaching), (2) pale (moderately bleached), or (3) severely bleached (white). Reef-wide bleaching prevalence for each species was determined as the proportion of observations of that species showing signs of moderate or severe bleaching (i.e. bleaching score of 2 or 3).