Part of the regional summaries: Philippines - Palawan and Philippines

One of Anderson's favorite Philippine dives, a structurally varied rocky fore-reef with striking genus diversity - branching and tabular Acropora, Seriatopora, Pocillopora, Montipora, Porites, brain corals, even cuttlefish, and a giant clam. Condition reads good, with isolated mortality spots and physical breakage rather than disease or bleaching. Cover estimate is around 27-31%.
One independent survey station shown with the program’s logo — click one to see how far it is from this site.
22 of these 25 photographs are shown beside the text that discusses them under "Site description" below.
It is a gently rolling fore-reef around an offshore rock outcrop, low to moderate relief, with sand and rubble channels at the deeper margins; the water is hazy and several frames are soft from a fogged housing, which he flags himself ("before fogging up", "finally unfogged") (e.g., Image 7, Image 14, Image 15).
Dense foliose, ruffled brown Turbinaria-type coral dominates one section with reef fish moving through it, and nearby a large plate coral with a pale growing margin sits amid purple encrusting sponge and yellow coralline algae - a structurally varied rocky reef (e.g., Image 2, Image 5).
Genus diversity is the site's signature: mixed branching and corymbose Acropora, tabular and shelf-like plates with clean pale margins, pink-to-mauve fine-branched stands he identifies as Seriatopora hystrix, his Pocillopora verrucosa and Pocillopora acuta, gold-tan digitate Acropora, tiered Montipora / Echinopora / Turbinaria whorls, his own Pavona meadow, massive and columnar Porites, Platygyra and Goniastrea maze colonies, fungiids, and a giant clam (e.g., Image 1, Image 3, Image 6, Image 9, Image 10, Image 11, Image 12, Image 13, Image 16, Image 17, Image 18, Image 22, Image 24, Image 25).
Cover was recovered by splitting the pooled 305-frame "bacuit_archipelago" computer vision collection back into its source folders - it merges all three El Nido dives with a separate 23 January island-hopping day, so its published row belongs to no pin. The 98 frames that are this dive give DINOv3 27.2 percent hard coral (1,960 points, 95 percent CI 25.3-29.0, frame range=8.4–58%) and CoralSCOP computer vision 31.3 percent over all 98 frames now scored (median=31.5, range=4.5–64%) - higher than the 28.6 percent its earlier six-frame sample gave. The two methods are also a check on the eye: an earlier visual read put this reef at 60-80 percent live cover, the models put it at 27 and 31, and the wide-angle frames show substantial turf, rubble, and standing dead framework between live colonies. They still fail differently - CoralSCOP computer vision reads high on low-cover reefs (median=1.68x a co-located diver survey, 4.6x at worst); DINOv3 is strong in-domain (6-class macro-F1 0.62) but degrades on unseen reefs to 0.17-0.25 cross-domain point accuracy. Neither is a diver survey; both are provisional. Coral cover: 29% — photo-derived, the mean of two computer vision methods. This site's figure is the mean of the DINOv3 calibrated point classifier reading over the 98 frames of this dive (27%) and the CoralSCOP computer vision figure over the same 98 frames (31%), a spread of 4.1 percentage points. The two products have different denominators and should not be treated as one measurement; the honest statement of what is known here is the 27–31% range, from the DINOv3 reading to the CoralSCOP run. Both are provisional photo-derived indices, not a diver survey;
Condition reads good with no bleaching front: the scattered stark-white patches on single colonies are isolated recent-mortality spots, and "brittle reef" is his own note on broken branching framework and rubble in the shallows - physical breakage, not disease (e.g., Image 4, Image 8).
No crown-of-thorns, no disease lesions.
Cuttlefish appear in five frames and at least two individuals, a crocodile flathead twice, plus anemonefish, butterflyfish, schooling yellow fusiliers, and crinoids (e.g., Image 23).
No independent survey was made at this exact site; the nearest ones are listed in the table below.
Paglugaban Rocks - the second Bacuit Archipelago dive of 24 January 2017, in at 12:29 for 58 minutes to a maximum of 20.1 m (mean 13.4 m) at 26.7 C, and the one Anderson called one of his favorite dives in the country.
All 98 frames were reviewed, every one of them underwater.
| Method | Hard-coral cover (%) |
|---|---|
| In-house DINOv3 | 27.2 |
| CoralSCOP | 31.3 |
| Mean (± SD) | 29.2 ± 2.9 |
| Range | 27.2–31.3 |
Mean ± SD and range are across the computer-vision methods shown.
Within 5 km, 1 independent survey station recorded hard coral cover in 4 surveys between 2013 and 2014. Of the 1 station surveyed in more than one year, 1 increased between their first and latest surveys. The longest record, Abdeen's Rock (Reef Check, 3.57 km away), went from 42% in 2013 to 61% in 2014. The most recent nearby survey is from 2014, so these values predate recent heat stress. This site's own estimate (29%) sits 12.3 points below the range of these surveys (42–61%); the methods differ, so the comparison is context, not a like-for-like test.
Collected by other programs using different methods (Reef Check uses point-intercept transects), so these are context for the figures above, not a like-for-like comparison. Where a station was surveyed more than once the range is shown. Station names are not dive-site names.
| Agency | Site / location | Date | Hard coral cover (%) | Distance | Link |
|---|---|---|---|---|---|
| Reef Check | Abdeen's Rock | 2013–2014 | 42–61% (4 surveys) | 3.57 km | — |
All images from Coral Reef Diagnostics. Data are from Coral Reef Diagnostics and the organizations named in the data-sources table above.
© Coral Reef Diagnostics.