Hard coral cover gave no clear linear link to bleaching report presence in four Atlantic and Pacific samples in 2003-2019

We tested a linear link between hard coral cover and the presence of a bleaching percentage in archived Reef Check surveys. Four site, year and month models used 7,059 repeated-site records in the Atlantic and Pacific in 2003-2010 and 2011-2019. Models controlled depth and surface heat stress. Slopes ranged from -3.42 to +1.34 reporting percentage points per 10 points of cover. All four adjusted working intervals included zero. Sensitivity checks did not establish a common link. A stricter later Atlantic sample had no missing bleaching values and was uninformative. These results do not establish random reporting or a small effect. Selected sites, unknown observers and visits, few year clusters, linear assumptions and near-complete reporting in one group limit inference. The package gives a derived panel, raw source hashes, code, all estimates, numerical checks and a figure. This is an archive observation test, not a causal climate effect.

coral bleaching · observation bias · missing data · hard coral cover · marine heat stress · Reef Check · null result