First-visit bleaching gave no clear common link to 90-day coral survey returns in four Atlantic and Pacific samples, 2003-2019

We tested whether bleaching at the first recorded coral survey in a year predicted a return on a distinct date within 90 days. Four Atlantic and Pacific models used 3,612 site-years in 2003-2019 and 201 recorded returns. Models controlled site, year, first month, survey-date heat, depth and sample count. Slopes ranged from -5.50 to +2.18 percentage points per 10-point increase in bleaching. 0 of four country-cluster intervals excluded zero after correction for four tests. These data did not establish a common return link. This is not evidence that visits were random. Sparse returns, unknown field visit logs, selected repeat sites, incomplete bleaching reports and uncertain dates limit inference. The result concerns the observation process in a heat-stress archive. It does not identify a causal climate or pollution effect. The package includes data extracts, code, source hashes, checks and a figure.

coral bleaching · heat stress · observation bias · recorded returns · regional replication · survey effort