### BITB41 — Three Brothers
**CRD page:** [Three Brothers/Central Chagos Coral Surveys BITB38-BITB49 (Mar 2015)](https://coralreefdiagnostics.com/three-brotherscentral-chagos-20150321-to-0324)

**22 Mar 2015**
**GPS:** 6.02997°S, 71.54900°E
**Setting:** intermediate-exposure submerged bank reef, 28 m (outside the lagoon; `exposure-original` reads "windward", `exposure-revised` "intermediate")
**Coral cover:** 41.9% (14 formal transects, D3–D5; **algal cover 28.8%**) · **published image bleaching (`Mean(%bleaching)` column):** 7.65%
**Live hard coral cover across methods:** formal transects 41.9% · CoralNet points 25.4% · Coralscapes masks 29.5% (13 frames). A fourth figure, the published image-cover product, gives 31.8% across 13 rows. These products have different sampling footprints and denominators and should not be averaged.
**Temperature:** 29.2 °C · **Salinity:** 34.7
**Photo review:** all 13 of 13 jpg stills examined (camera timestamps 08:43:54–09:23:54, a single continuous dive)
**Representative photos:** P3220387.jpg, P3220391.jpg, P3220400.jpg, P3220415.jpg, P3220425.jpg

BITB41 is a submerged bank reef at 28 m on the intermediate-exposure side of Three Brothers, and it carries **41.9% coral against 28.8% algae — 46th of 109 Chagos sites for coral cover, and 90th of 109 for algal cover, i.e. in the lowest fifth of the archipelago for algae.** That low algal figure is not the usual Chagos story of a hard, coralline-paved reef. **Crustose coralline algae here is only 8.6% of the benthos (88th of 109) and turf only 6.6% (86th of 109)** — both near the bottom of the archipelago — while erect algae runs to 7.3% and macroalgae to 3.9%, unusually high shares of a small algal budget, with sediment-laden turf at 2.3% and cyanobacteria at 0.06%. What takes up the space instead is **bare substrate: 15.4% of the benthos**, plus 13.9% non-coral invertebrates. This is a patchy reef whose non-coral ground is open sand and rubble rather than algal pavement, and the photographs show exactly that.

All genus values below are **percent of total benthos, not percent of coral**, and the 28 nonzero genera sum to 41.944% — an exact match to the published cover figure. **The assemblage is overwhelmingly *Porites*-led: 17.96% of the benthos, which is 43% of every coral point recorded at the site**, and more than four times the next genus. *Pachyseris* follows at 3.93%, then *Montipora* 3.43%, *Platygyra* 2.57%, *Pocillopora* 2.45% and *Acropora* 2.03%. **That *Acropora* figure is the notable one — 4.8% of the coral present, sixth in rank, on a reef where it would normally lead** — and it is consistent with the depth (28 m) and with an imagery set in which tabular and corymbose *Acropora* appear but never dominate a frame. Behind the leaders sit *Pavona* 1.22%, *Favia* 1.20%, *Favites* 1.19%, *Echinophyllia* 1.10%, *Stylophora* 0.83%, *Psammocora* 0.74%, *Leptastrea* 0.71%, *Turbinaria* 0.53% and *Montastrea* 0.51%, then a tail of 13 genera under 0.3% each — *Cantharellus*, *Astreopora*, *Symphyllia*, *Cyphastrea*, *Fungia*, *Goniastrea*, *Echinopora*, *Physogyra*, *Oxypora*, *Gardineroseris*, *Leptoseris*, *Acanthastrea* and *Hydnophora*. **Genus count is 28, 29th of 109 Chagos sites** — upper quartile, though rare codes at the bottom of that list may rest on only a few survey points. Diversity indices computed on the 28 codes give Shannon H′ = 2.20, evenness J′ = 0.66 and an effective diversity of 9.05 genera, i.e. moderate evenness held down by the *Porites* dominance.

**The three depth strata look almost identical in cover — and that flatness is itself the finding, because it is not matched by compositional flatness.** D3 gives 44.3% coral against 32.2% algae (n = 10 transects), D4 gives 40.5% against 34.2% (n = 3), and D5 gives 41.0% against 20.0% (n = 1). Cover therefore varies by under 4 pp across a 3-stratum profile, but the individual transects range from 28% to 73% coral, so **local patchiness within a stratum is roughly five times the difference between strata** and no depth trend in cover can be defended from these data. Composition, by contrast, does change. *Porites* leads everywhere (19.76% at D3, 15.13% at D4, 19.00% at D5) but the supporting cast turns over completely: *Acropora* declines from 3.68% at D3 to 2.40% at D4 to **zero at D5**, while ***Pachyseris* does the reverse — absent at D3, 4.80% at D4, 7.00% at D5**, and *Platygyra* climbs from 1.18% to 1.53% to 5.00%. *Montipora* peaks in the middle (3.90% → 6.40% → 0%). Genus count collapses down the profile as well: 26 genera at D3, 23 at D4, **8 at D5** — but D5 is a single 100-point transect, so that is a sampling floor, not an ecological one. Bare substrate rises steeply with depth (5.8% → 15.3% → 25.0%) while algal cover falls (32.2% → 34.2% → 20.0%), which is the clearest real depth signal in the dataset: the deep stratum is where the sand and rubble lanes take over.

The photographs document that patch-and-lane architecture directly. **P3220386, P3220388, P3220423, P3220424 and P3220425 are all substantially open ground** — broad white carbonate sand and rubble plains with low colonies scattered across them, and in P3220388 the coral is reduced to isolated plates and small heads on an almost continuous sand sheet. Against these, P3220400, P3220401 and P3220416 show the coral-abundant face of the same reef: near-continuous mosaics of tabular and plating colonies, ridged foliose colonies consistent with *Pachyseris*, brown scroll-shaped *Turbinaria* cups, ochre digitate *Millepora*-type colonies, corymbose *Acropora*, massive *Porites* and hemispherical meandroid heads consistent with *Platygyra*. **P3220391 is the structural set-piece: a large hemispherical brain colony sitting directly on a broad olive-yellow encrusting sheet, with a clean white-and-pink tissue-loss margin and a dark necrotic band along the whole contact zone — a textbook coral-versus-coral overgrowth boundary**, with damselfish, a butterflyfish and dark surgeonfish working the sheet around it. **P3220415 confirms the site's unusual algal budget**: a dense, clearly identifiable *Halimeda*-type patch with white calcified segments growing in the gap between a ridged *Pachyseris*-compatible colony and a smooth *Porites*-compatible one — the photographic counterpart to 7.3% erect algae plus 3.9% macroalgae on a reef with almost no turf or CCA. **P3220390 is a puzzle worth naming: a large domed colony with dense, uniformly extended club-tipped polyps, consistent with *Goniopora* or *Alveopora* or with a corallimorpharian aggregation — and neither *Goniopora* nor *Alveopora* is recorded anywhere in the BITB41 transect data**, so either the taxon was missed by the transects or the animal belongs in the 13.9% non-coral invertebrate class.

**Condition reads as localized paling rather than a bleaching event, and the photographs are unambiguous on that point.** `Mean(%bleaching)` is **7.65%, 45th of the 98 Chagos sites carrying a value — squarely mid-table for bleaching severity**, and the two CoralNet training runs agree here (7.65 vs 7.653), so this is not one of the ten sites where the runs diverge. Bleaching at this site depends heavily on the denominator: the published product reports 7.65%, while 14 paired bleached points work out to 3.6% of all 390 annotated points, or 12.4% of the 113 coral points — three legitimate numbers, three different questions. **The imagery supports real but patchy paling. P3220387 shows a stark-white branching colony of *Pocillopora*/*Stylophora*-compatible form, part of it still white-with-tissue and part already overgrown by yellow-green filamentous algae, growing against a large blue-grey giant clam mantle** — a previous reviewer left this organism's identity unresolved; at frame resolution the corallite texture is visible and it reads as a bleached-to-recently-dead branching scleractinian rather than a soft coral or sponge. **P3220389 shows an olive-brown *Pocillopora*-type colony with numerous stark-white branch tips against otherwise pigmented tissue** — classic partial tip-paling on an otherwise healthy colony. Neither frame, nor any other in the set, shows a coherent site-wide bleaching field: the great majority of colonies across all 13 frames carry normal olive, tan, purple and ochre pigmentation. **No COTS individuals or feeding scars, no disease front, no *Drupella* aggregation and no clear corallivore scarring were observed** in the full 13-frame review — the necrotic margin in P3220391 is a competitive interaction, not a disease lesion.

Fauna in the imagery is present but not the dive's highlight. Two yellow-tailed blue fusiliers (*Caesio*-type) cross P3220400 along with a larger barred fish at the frame edge and a bannerfish; damselfish, a butterflyfish and a small school of dark surgeonfish work the *Porites* sheet in P3220391; scattered chromis appear in P3220416 and P3220423. A large giant clam anchors P3220387. A previous reviewer logged a probable camouflaged octopus in P3220425 at medium confidence; this pass could neither confirm nor exclude it — the candidate is a mottled maroon-and-olive mound of the right size and texture, but it is equally consistent with an encrusting-sponge-covered rubble block.

<b>Assemblage composition.</b> Formal surveys give 41.9% coral, 28.8% algae, 15.4% bare substrate and 13.9% non-coral invertebrates across **28 nonzero genera (29th of 109 Chagos sites)**, with *Porites* alone holding 17.96% of the benthos — 43% of all coral present — and *Acropora* an unusually minor 2.03%. The three depth strata are near-identical in cover (44.3 / 40.5 / 41.0%) but not in composition: *Acropora* falls to zero and *Pachyseris* rises from zero to 7.00% between D3 and D5, while bare substrate climbs from 5.8% to 25.0%. The photographs document both faces of the reef — dense plate-and-massive mosaics in P3220400/401/416 against broad sand-and-rubble lanes in P3220386/388/423/425 — plus a *Halimeda* patch consistent with the site's unusual erect-algae budget, a coral-on-coral overgrowth margin in P3220391, and localized bleached tips and one bleached-to-dying branching colony, with no COTS, disease front or site-wide bleaching field anywhere in the 13 stills.

**Flags:** (1) **D5 rests on a single transect** (`BITB41_d5_T1`, 100 points) yet contributes a full third of the headline cover figure, because the site value is the unweighted mean of the three stratum means. Every D5 statement above — 41.0% cover, 8 genera, *Pachyseris* at 7.00%, *Acropora* at zero, 25.0% bare substrate — rests on that one transect, and D5's 8 genera should not be read as a real deep-water diversity collapse. D4 is little better at n = 3. (2) ***Pachyseris*'s site-level 3.93% is generated entirely by the two thin strata.** It records zero at D3 (n = 10) and takes its whole site value from D4 (n = 3) and D5 (n = 1). Treat the site figure as provisional. (3) **The transects mix 100-point and 500-point totals** and are compared through percentage fields; the site value averages stratum means rather than weighting all 14 transects equally. (4) **Bleaching is quoted three ways with three denominators** — 7.65% published, 3.6% of all 390 annotated points, 12.4% of 113 coral points. They are not interchangeable, and the shard states which is which every time. (5) **Two frames show genuine paling.** P3220387 carries a stark-white branching colony, partly algal-overgrown, adjacent to a giant clam — reidentified in this pass from "unresolved organism" to a bleached-to-recently-dead branching scleractinian; P3220389 shows white branch tips on an otherwise pigmented *Pocillopora*-type colony. These support localized paling, not a site-wide bleaching field. (6) **P3220390 shows a taxon absent from the transect data** — a domed colony with dense club-tipped polyps consistent with *Goniopora*/*Alveopora* or a corallimorph aggregation, neither genus recorded at BITB41. Either a transect miss or a non-coral invertebrate; unresolved from a single still. (7) **The probable octopus in P3220425 remains medium confidence** and could not be confirmed on re-review. (8) **No COTS, disease front or corallivore scarring was observed** across the full 13-frame review — an explicit absence-of-evidence finding. The white-and-pink margin in P3220391 is coral-on-coral competition, not disease. (9) **No per-frame depth assignments exist** for the photographs, so imagery cannot be matched to a stratum, and there are **no D1–D2 transects at this site** — the shallow reef is simply unsampled. (10) **The site carries four different live-cover numbers** (41.9% transect, 31.8% published image product over 13 rows, 29.5% Coralscapes masks, 25.4% CoralNet points; `n_photos=13`, matching the folder exactly). The transect value is ground truth for cover; the CV products sit 12.4–16.5 pp below it and must not be averaged with it or with each other. (11) **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`). The site table's `Mean(%bleaching)` column is not a mean — it equals the 1st CoralNet training run — so any rank taken off it is approximate; it also carries at least one known bad value elsewhere in the archipelago (BITB58 listed at 22.22% against a corrected 0.77%). Never read this number as a diver's field estimate.

**Benthic data:** [Download BITB41 benthic survey data (CSV)](../benthic_data/BITB41_benthic.csv) — scleractinian cover by genus with full names, plus substrate and algal categories (bare substrate, CCA, turf, macroalgae, invertebrates), from the LOF BIOT-1 survey (Mayfield & Dempsey 2026, *Oceans*, supplemental file S1).