### BITB55 — Three Brothers
**CRD page:** [Three Brothers/Central Chagos Coral Surveys March 25-28, 2015](https://coralreefdiagnostics.com/three-brotherscentral-chagos-20150325-to-0328)

**26 Mar 2015**
**GPS:** 6.17526°S, 71.53974°E
**Setting:** intermediate-exposure, emergent fringing reef (outer forereef, South Brother), 29.6 m
**Coral cover:** 47.6% (3 formal transects, D2–D4; **algal cover 24.4%**) · **published image bleaching:** 8.12%
**Live hard coral cover across methods:** formal transects 47.6% · CoralNet points 35.6% · Coralscapes masks 29.9% (13 frames). These products have different sampling footprints and denominators and should not be averaged.
**Temperature:** 30 °C · **Salinity:** 34.3
**Photo review:** all 29 photographic frames examined (the folder also holds 3 video clips — P3260421.MOV, P3260422.MOV, P3260443.MOV — not reviewed here)
**Representative photos:** P3260439.jpg, P3260434.jpg, P3260409.jpg, P3260410.jpg, P3260419.jpg

BITB55 is a coral-abundant forereef off South Brother, with **47.6% transect coral cover against 24.4% algae** and only 23 nonzero genera. The benthos is otherwise tidy: bare pavement 16.2%, non-coral invertebrates 11.8%, crustose coralline algae 9.1%, turf 8.2%, sediment-laden turf 5.6%, macroalgae 1.2% and erect algae 0.3% — together with the coral and algal figures these sum to essentially 100% of the recorded benthos.

The assemblage is carried by two genera. ***Porites* holds 15.83% of the benthos and *Acropora* 10.87%** — together 56% of all coral recorded — with *Montipora* third at 5.53% and *Pocillopora* fourth at 2.97%. Then *Pavona* 2.25%, *Pachyseris* 1.67%, *Astreopora* 1.48%, *Favia* 1.36%, *Leptastrea* 1.12% and *Favites* 0.85%, with *Cyphastrea*, *Psammocora*, *Platygyra*, *Goniastrea*, *Gardineroseris*, *Echinophyllia*, *Leptoseris*, *Millepora*, *Echinopora*, *Montastrea*, *Galaxea*, *Symphyllia* and *Turbinaria* filling out the tail below 0.65%. All of these are percentages of total benthos, not of coral alone; the 23 genus values sum to exactly 47.60%, matching the published site-level cover with no residual.

**The depth structure is a genus handover from *Acropora* to *Porites*, and it is the most informative single result at this site.** D2, the shallowest transect, carries 49.8% coral of which *Acropora* is 23.60% — 47% of the coral on that transect, close to a co-dominant — with *Porites* second at 8.48% and the reef's highest algal cover (32.96%) and richest genus list (19 codes). D3 holds the site's highest cover at 57% and flips the order: *Porites* 17.00% now leads, *Montipora* rises to second (9.75%), and *Acropora* drops to third (9.00%, just 16% of the coral on that transect) — still 17 genera present. D4, the deepest stratum, holds 36% coral and is close to a *Porites* monoculture: **22.00% of the benthos, 61% of all coral on that transect**, with *Pachyseris* a distant second (5.00%) and *Acropora* absent from the transect altogether. Genus richness collapses from 19 (D2) to 17 (D3) to just 8 (D4), and D4 also carries the lowest algal cover (10%) of the three strata. **The site-level figures of 15.83% *Porites* and 10.87% *Acropora* are an average of a shallow *Acropora*-and-*Porites* mix and a deep near-*Porites* monoculture; neither number describes D2 or D4 individually.**

The photographs document both halves of that structure. The shallow, *Acropora*-rich zone is spectacular: P3260439, P3260440, P3260441, P3260442 and P3260444 show extensive tabular *Acropora* fields — overlapping pedestalled tables running to the edge of visibility, interspersed with corymbose and digitate thickets and the occasional bright green or purple colony, with a diver visible for scale in P3260440. The deeper, *Porites*-heavier zone is represented by large tan-cream massive and lobate domes in P3260409, P3260411, P3260426, P3260434, P3260436, P3260437 and P3260438, several photographed at close range by a diver with an underwater housing (P3260411, P3260426). Dark purple-brown digitate thickets consistent with *Pocillopora* appear threaded through several of the same frames (P3260415, P3260434, P3260436, P3260437, P3260442). A set of close macro frames (P3260403, P3260404, P3260406, P3260407) show large cerioid/meandroid colonies with fluorescent turquoise polyp centers, consistent with a faviid such as *Favites* or *Platygyra* — genera the transects record at low but nonzero cover (0.85% and 0.58% respectively). P3260392 shows a comparatively sparse reef-wall face — mostly bare pavement and coralline crust with scattered foliose and encrusting colonies, a school of surgeonfish, and a soft-coral (gorgonian) whip — the visual counterpart of the site's 16.2% bare substrate and 9.1% CCA.

No colony in any of the 29 frames shows the stark, colony-scale white pallor that would flag active bleaching; all corals photographed, from the tabular *Acropora* to the massive *Porites*-type domes, are normally pigmented in tan, brown, olive, purple, cream or green. This is consistent with the modest published bleaching figure of 8.12% and gives no reason to doubt it.

Fauna is a notable component of this folder — its name ("octoporn") is not an exaggeration. A single large octopus, pale grey-blue and mottled when at rest and darker and more actively patterned when alert, is documented resting in and emerging from a coral-rubble den across seven of the 29 frames (P3260410, P3260412, P3260413, P3260414, P3260418, P3260423, P3260424); its size and pale resting coloration are consistent with the day octopus, *Octopus cyanea*, common on Indo-Pacific reef flats, though it is not confirmed to species. A small orange nudibranch with white-ringed black tubercles sits on coralline rock in P3260401. Most notably, P3260419 shows a diver using surgical forceps to lift a small multi-armed sea star from the reef — its spination and disc pattern are consistent with a juvenile crown-of-thorns sea star (*Acanthaster*-type, not confirmed to species) — evidently a deliberate in-water check or removal during the survey rather than an incidental sighting.

<b>Assemblage composition.</b> Formal transects give 47.6% coral and 24.4% algae across three depth strata, with the coral split roughly evenly between *Porites* (15.83% of benthos) and *Acropora* (10.87%) at the site level — an average that conceals a real depth handover from an *Acropora*-*Porites* mix in the shallows to a near-*Porites* monoculture at depth. The photographs independently document both ends of that gradient (tabular *Acropora* fields shallow, massive *Porites*-type domes deep) plus a faviid-type meandroid component, a resident octopus, and one direct crown-of-thorns observation.

**Flags:**
1. **Depth zonation is an *Acropora*-to-*Porites* handover, not a stable community.** D2 (shallow) carries *Acropora* at 47% of its coral; D4 (deep) is 61% *Porites* with *Acropora* absent. Never quote the site-level 15.83% *Porites* / 10.87% *Acropora* as describing any single transect.
2. **Genus richness collapses with depth** (19 genera at D2, 17 at D3, 8 at D4), alongside declining algal cover (32.96% → 30.25% → 10%). Consistent with a shallow high-light, high-turbulence *Acropora* zone giving way to a shaded, lower-diversity *Porites* bank.
3. **Crown-of-thorns sea star handled by the diver in P3260419** — a small individual, spination/disc pattern consistent with *Acanthaster* (not confirmed to species), being lifted from the substrate with forceps. This reads as a deliberate in-water check/removal rather than a passive sighting; worth confirming against any expedition COTS-control log.
4. **Large resident octopus photographed across 7 of 29 frames** (P3260410, 412, 413, 414, 418, 423, 424), consistent with *Octopus cyanea* but not confirmed to species — explains the folder's "octoporn" name; these are repeat observations of one animal, not independent sightings.
5. **Multi-method frame-count mismatch.** The photo folder holds 29 still frames (plus 3 unreviewed video clips), but `chagos_multimethod_cover_by_site.csv` records `n_photos = 13` for the CoralNet/Coralscapes comparison — fewer than half the available stills. Treat the 35.6%/29.9% method figures as resting on a partial, unspecified subset of the imagery reviewed here.
6. **Point annotation set is majority machine-derived.** Per site-wide provenance, roughly three-quarters of the underlying CoralNet point annotations across the Chagos corpus are unconfirmed robot predictions rather than human-confirmed calls; nothing here should be read as "the annotator observed," only as the point set's output.
7. **Integrity check passes:** genus columns sum to 47.60% against the stated 47.60% coral cover at the site level, and to 49.8/57/36% against the three stated depth-row cover values, all exact.
8. **The published bleaching figure is a COMPUTER-VISION product, not an in-situ observation.** It comes from CoralNet annotation of the informal survey images (Mayfield & Dempsey 2026, *Oceans*). Per-image human-vs-model agreement is substantially weaker for bleaching than for cover: r = 0.40 for bleached against r = 0.72 for live hard coral (`CHAGOS_MULTIMETHOD_README.md`). Treat the site bleaching percentage as the weaker of the two published numbers, and never read it as a diver's field estimate.

**Benthic data:** [Download BITB55 benthic survey data (CSV)](../benthic_data/BITB55_benthic.csv) — scleractinian cover by genus with full names, plus substrate and algal categories, from the LOF BIOT-1 survey (Mayfield & Dempsey 2026, *Oceans*, supplemental file S1).