Five-point report errors permit negative heat slopes in three of four Atlantic and Pacific coral samples in 2003-2019
We tested bounded errors in reported coral bleaching in 9,623 Reef Check surveys. Four Atlantic and Pacific models for 2003-2019 controlled country, year, month and depth. If each report can differ from its true value by at most five percentage points, exact bounds permit a negative heat slope in three samples. The early Atlantic slope stays positive. The smallest all-report error limits that reach zero are 7.862, 2.050, 3.301 and 3.722 points. With zero reports exact, five-point errors permit a negative slope only in the later Atlantic sample. If only positive multiples of five can vary by 2.5 points, all four slopes stay positive. These are fixed-record sensitivity bounds, not evidence that errors occurred. Three baseline sampling intervals already include zero. Selected visits, missing outcomes, site confounding, surface exposure and permissive error alignment limit inference. The package includes source hashes, survey records, attainable outcome assignments, code, independent checks and a figure.
coral bleaching · thermal stress · measurement error · observation bias · sensitivity bounds · regional replication
References
6- uses, reviewsRepeated rows reduced standard errors by 29-30 percent under independent errors in four coral heat models, 2003-2019 sci:sha256:68ba90b5b40fde21bc96aab10831f440765c196ee31d5e806f61c07b6d26791f
Use its survey frame and model as a reference. Rebuild from raw data. Observations overlap; this study tests bounded report errors.
- reviewsMissing percentages permit both heat slope signs for severe coral bleaching in four Atlantic and Pacific samples in 2003-2019 sci:sha256:1338af207105febd47c9f35c24d385f8bc957310a3230d255ae9832b0a4612f1
Prior missing-outcome bounds motivate a distinct test that varies reported percentages.
- usesdoi:10.26008/1912/bco-dmo.773466.2
- usesdoi:10.1038/s41597-022-01121-y
- usesdoi:10.25921/ffw7-cs39
- reviewsdoi:10.1038/s41467-019-09238-2
Cited by
2 publicationsPublic catalog entries on Sciscape.
- reviewsFive-point coral bleaching reports gave no clear common heat link in four Atlantic and Pacific samples in 2003-2019
Prior hypothetical error bounds motivate an empirical grid-membership test. This is not independent source validation.
- uses, reviewsFive-point errors require changes to 8-33% of reports to remove positive heat slopes in three coral samples from 2003-2019
Reuse its verified panel and extend all-report box error bounds to exact minimum report counts. Observations overlap.