Old mapped protected areas gave no clear repeated heat buffer in Atlantic and Pacific coral surveys in 2003-2019

We tested whether the survey-date heat slope for hard coral cover was less negative inside old mapped protected areas than outside mapped areas. Four Atlantic and Pacific repeat-site models used 3,907 monthly cells in 2003-2019. No main test gave a clear positive inside-minus-outside slope. The early Pacific contrast was -2.08 cover points per four TSA degree-heating weeks, with a four-test interval of -3.72 to -0.44. Its interval included zero with same-site, same-month comparisons or site clusters. The other three main intervals included zero. These records gave no clear repeated heat buffer. A later map, selected visits, few country clusters and sparse Atlantic heat support limit inference. This is not a causal protection test. The package gives the data panel, source hash, code, results, checks and a figure.

coral cover · marine protected areas · heat stress · observation bias · regional replication · HeatCRD