Heat gave no clear link to survey depth in four Atlantic and Pacific coral samples in 2003-2019

We tested whether coral survey depth changed with surface heat dose at the same site and calendar month. We used 3,288 repeated Reef Check site-month records in the Atlantic and Pacific in 2003-2010 and 2011-2019. Separate models controlled site-calendar-month and year. Mean depth slopes were +0.193, +0.184, +0.031 and -0.102 metres per four TSA_DHW units. All four region-cluster intervals included zero after correction for four tests. Both Pacific intervals stayed within -1 to +1 metre. The later Atlantic interval was -1.723 to +2.091 metres. Thus the records give no clear directional heat link to survey depth, with tighter bounds in the Pacific. This does not rule out selection of sites, dates or observers, nonlinear depth changes, or depth-related detection error. Few Atlantic regions, limited high-heat data, coarse surface heat and selected repeat visits limit inference. The package gives a source hash, data panel, code, checks, results and a figure.

coral surveys · marine heat stress · survey depth · observation bias · Reef Check · regional replication · null result