### BICC50 — Central Chagos (photo archive code BITB50)
**CRD page:** [Three Brothers / Central Chagos Coral Surveys (25–28 Mar 2015)](https://coralreefdiagnostics.com/three-brotherscentral-chagos-20150325-to-0328)

**25 Mar 2015**
**GPS:** 6.39945°S, 71.61726°E
**Setting:** intermediate-exposure submerged bank reef, inside the lagoon, 20.0 m
**Coral cover:** 12.1% (2 depth strata, D3–D4, 11 transects; **algal cover 57.9%**) · **published image bleaching:** 24.2%
**Live hard coral cover across methods:** formal transects 12.1% · CoralNet points 13.9% · Coralscapes masks 12.8% (11 frames). These products have different sampling footprints and denominators and should not be averaged.
**Temperature:** 30.0 °C · **Salinity:** 34.2
**Photo review:** all 11 of 11 frames examined (folder `BiTb50-finished`)
**Representative photos:** P3250389.jpg, P3250406.jpg, P3250405.jpg, P3250408.jpg, P3250427.jpg

BICC50 is a submerged bank reef inside the Central Chagos lagoon, surveyed at 20.0 m under intermediate exposure, and **it is one of the lowest-cover sites in the archipelago: 12.1% coral against 57.9% algae, ranking 102nd of 109 Chagos sites for coral cover and 10th of 109 for algal cover.** The rest of the benthos is almost entirely unconsolidated or algal-veneered ground: bare substrate accounts for 24.3% of the benthos and non-coral invertebrates 5.8%. Within the algal budget, **sediment-laden turf (PTSTS) alone is 33.2% of total benthos and erect algae (PETS) a further 14.6%**, while crustose coralline algae is almost absent at 2.9% (108th of 109 Chagos sites — second-lowest in the archipelago) and clean fleshy turf is 4.2% (98th of 109), with macroalgae at 3.0%. The five algal categories sum exactly to the published 57.92%. This is not a well-grazed coralline pavement; it is a sediment-draped, algae-dominated bank with sparse coral on it, and the photographs match that description closely.

The coral that is present is spread thinly across few taxa. **Fourteen nonzero genera are recorded — 96th of 109 for genus richness** — and the fourteen genus percentages sum to 12.0542% of the benthos, an exact match to the published cover figure. All values below are percent of total benthos, not percent of coral. *Acropora* leads at 4.97% of the benthos (41% of all the coral at the site), then *Montipora* 2.59% (21%) and *Porites* 2.29% (19%); **those three genera hold 82% of the coral present.** *Turbinaria* follows at 0.56%, then *Pocillopora* 0.36%, the hydrocoral *Millepora* 0.35%, *Lobophyllia* 0.25%, *Echinophyllia* 0.19%, *Leptastrea* 0.18%, *Fungia* and *Favites* 0.09% each, *Goniopora* 0.08%, *Coscinaraea* 0.06% and *Galaxea* 0.01%. In absolute terms even the leading genus covers under 5% of the benthos, so no genus here is abundant — the ranking describes which taxa are least scarce.

**The depth structure is inverted relative to most Chagos forereefs: the deeper stratum carries three and a half times the coral of the shallower one.** Only two depth strata were surveyed, D3 and D4, and the replication is uneven — **11 transects in total, 3 at D3 and 8 at D4.** D3 carries just **5.33% coral against 62.33% algae**, with *Acropora* 2.33% (44% of that stratum's coral), *Montipora* 1.00%, *Porites* 1.00%, *Millepora* 0.67% and *Leptastrea* 0.33% — a stratum where live coral is a minor component of a predominantly algal and sediment-covered surface. D4 rises to **18.78% coral against 53.50% algae**, with *Acropora* 7.60% (40% of that stratum's coral), *Montipora* 4.17%, *Porites* 3.58%, *Turbinaria* 1.12%, *Pocillopora* 0.72% and *Lobophyllia* 0.50%. The genus order is essentially unchanged between strata — the difference is one of amount, not of composition. **The site-level 12.05% figure is the equal mean of the two stratum means, not the mean of the 11 individual transects**, and it therefore weights three D3 transects as heavily as eight D4 transects; the unweighted transect mean is higher. Note also that at D3 coral and algae together leave 32.3 pp of the benthos unaccounted for, and at D4 27.7 pp — consistent with the very high site-level bare-substrate reading, which is not itemised by depth in this dataset.

The photographs — all 11 frames in the folder — divide into seven wide habitat frames and four macro frames of a single subject, and they corroborate the transect picture directly. **P3250389, P3250406, P3250407, P3250408 and P3250409 all show the same low-relief bank: a broad, almost flat plain of pale sand and rubble veneered with dark filamentous and sediment-trapping turf, dotted with pink-mauve algal tufts and small nodules, and carrying only small, widely spaced coral colonies** — mostly low foliose or encrusting plates and occasional small mounds, with no framework-building thicket anywhere in the set and no vertical relief beyond a few tens of centimetres. A diver silhouette is visible at the top left of P3250389. P3250405 is the one frame that shows the coral at close range: a cluster of dark grey-brown ridged foliose plates alongside a green massive colony and an olive branching colony, growing directly out of coarse pale sediment; the ridged plates are consistent with *Turbinaria*, *Echinophyllia* or *Pavona*-type growth but image-based genus calls are tentative and none is asserted here. P3250419 shows a folded, lobed sessile colony with **two dark, pale-speckled molluscs on its surface** and an orange encrusting sponge alongside — the animals' identity is not resolved, and an earlier confident nudibranch call for this frame was withdrawn. **P3250426, P3250427, P3250430 and P3250431 are four consecutive macro frames of the same pair of gastropods with white, dark-spotted mantles fully extended over a smooth-textured coral surface** — a striking natural-history record, and, as set out in the Flags, the probable source of this site's very high published bleaching figure. Fauna otherwise is limited: **two shark passes are documented over the bank in P3250407, P3250408 and P3250409**, recorded in the earlier review of this folder as whitetip reef sharks, though fin-tip markings are not resolvable at this image resolution.

**Condition cannot be read straightforwardly from the published bleaching figure at this site, and the figure should be treated as unsafe.** The published image bleaching value is **24.2%, which ranks 3rd of the 98 Chagos sites carrying a value** — apparently one of the most bleached reefs in the archipelago. **The photographic evidence does not support that reading.** The seven wide frames show no field of white colonies and no colony-scale pallor beyond ordinary pale-sediment and dead-substrate tones; and an earlier audit of the point annotations for this folder found that **25 of the 26 bleached point labels fall in the four close-up gastropod frames (P3250426, P3250427, P3250430, P3250431), where the molluscs' white mantles and pale backgrounds were treated as coral**, while the Coralscapes masks report essentially zero bleached area. The 24.2% is retained here as published, because the published column is the citable source and is not recalculated in this workflow, **but it should be regarded as annotation-contaminated at this site and must not be used in cross-site bleaching comparisons without that caveat.** No disease lesions, tissue-loss margins, *Drupella* aggregations or predation scarring were identified in any frame, and no crown-of-thorns starfish were seen — though a set this small and this macro-heavy cannot establish true absence.

<b>Assemblage composition.</b> Formal transects give 12.1% coral against 57.9% algae — the algae being overwhelmingly sediment-laden turf (33.2% of benthos) and erect algae (14.6%), with coralline algae all but absent — plus 24.3% bare substrate, across only 14 nonzero genera of which *Acropora* (5.0%), *Montipora* (2.6%) and *Porites* (2.3%) hold 82% of the coral. The two depth strata run 5.33% cover at D3 (n=3 transects) and 18.78% at D4 (n=8), with the same genus order at both. The 11 photographs show exactly this: a flat, sediment- and turf-veneered lagoonal bank with small, widely spaced coral colonies, two passing sharks, and four macro frames of a pair of mantle-extended gastropods that are the likely source of the site's misleadingly high published bleaching value.

**Flags:** (1) **SITE IDENTITY — the photo archive calls this folder BITB50; the formal LOF dataset carries the same survey as BICC50.** Folders resolve to sites by number, not prefix (CC = TB = Three Brothers / Central Chagos). Both codes refer to one site: Central Chagos, 25 Mar 2015, 6.39945°S / 71.61726°E. **An older shard, `shards/desc_BITB50.md`, already exists for this same site in a pre-2026-08-26 format** (no ranks, no multi-method header line, no photo-review line); this file supersedes it and the duplicate should be resolved before any push to `crd_dive_sites.json`, so that one site does not get two entries. (2) **BLEACHING FIGURE IS UNSAFE AT THIS SITE — do not use in cross-site comparisons without this caveat.** The published 24.2% ranks 3rd of 98, but 25 of the 26 bleached point labels come from the four close-up gastropod frames (P3250426, P3250427, P3250430, P3250431), the Coralscapes masks give essentially zero bleached area, and the seven wide frames show no broad bleaching field. (3) **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`). (4) **Any rank taken off the `Mean(%bleaching)` column is approximate** — that column carries at least one known bad value (BITB58 listed at 22.22% against a corrected 0.77%), and this site's 3rd-of-98 placement is a second reason to distrust the top of that ordering. (5) **The multi-method row is keyed under BITB50 and its `transect_pct` cell is BLANK, with the note "no transect cover" and `n_methods` = 2.** That is a keying artefact, not a data gap: BICC50 has a published transect cover of 12.05%. The row should be re-keyed to BICC50 and the transect value filled in. (6) **With that correction, this is one of the rare Chagos sites where the CV products and the transects agree**: CoralNet points 13.9% and Coralscapes masks 12.8% against transects 12.1%, a spread of under 2 pp. This is consistent with the documented behaviour of the photo methods, which over-read by about +6 pp at 0–10% transect cover and under-read badly at 40%+ (`CHAGOS_MULTIMETHOD_README.md`) — agreement here is a property of the low-cover regime, not evidence that the methods are interchangeable. Still do not average them. (7) **`coralnet_points_pct` is documented as a human 30-point read, not a model prediction**, but only roughly a quarter of Chagos image annotations corpus-wide are Anderson-confirmed; and at this site the bleached labels are demonstrably wrong, so treat the whole annotation set for this folder with caution. (8) **Site-level cover is the equal mean of the two stratum means, not the mean of the 11 transects.** Replication is uneven (D3 n=3, D4 n=8) and D3 is thin; the unweighted transect mean is higher than 12.05%. (9) **No D1 or D2 transect exists** — the shallowest reef zones are unsampled and the profile starts at D3, so the site-level figure describes only the deeper part of the bank. The depth codes carry no metre boundaries; the 20.0 m in the header is the site-level survey depth. (10) **Bare substrate is very high at 24.3% of the benthos, and coral plus algae leave 27.7–32.3 pp unaccounted for at each depth stratum** — the depth-level summary does not itemise bare substrate or non-coral invertebrates. (11) **CCA is second-lowest in the archipelago (2.9%, 108th of 109)** while sediment-laden turf runs to 33.2% — an algal budget worth carrying into any cross-site comparison, and a signature of low water motion plus sediment loading on a lagoonal bank. (12) **The identity of the two dark speckled molluscs in P3250419 is unresolved**; an earlier confident nudibranch species call was withdrawn and is not repeated. The pair of mantle-extended gastropods in P3250426–431 is recorded as cowrie-compatible without a species call. (13) **Two shark passes (P3250407–409)** were logged in the earlier review of this folder as whitetip reef sharks; the identification is not independently confirmable at this image resolution. (14) **No COTS, disease or predation scarring was observed** across the full 11-frame review — but four of those frames are macro shots of a single subject, so the effective habitat sample is seven wide frames and cannot establish true absence. (15) **TOOLING — `sitefacts.py` reports "transect rows: 0" for this site (and, on inspection, for others).** `Chagos benthic data-all 1556 transects.csv` keys its first column `site x depth x transect` (e.g. `BICC50_d4_T3`), not `site code`, so the script's transect lookup never matches. The 11 transect rows quoted above were read directly from that file. (16) **HOUSEKEEPING — `chagos_site_to_page.json` has no entry for BICC50** (only for BITB50). The URL above is taken from the BITB50 entry; the mapping file should gain a BICC50 key. (17) **NON-CANONICAL BLEACHING RANK — added 2026-08-26.** Any "of 98" bleaching rank above was computed off the site table's `Mean(%bleaching)` column, which equals the **1st** CoralNet training run. Anderson's decision of 2026-08-26 (HOW_TO rule 4e) makes **`% bleaching (2nd training)` canonical**, The file carries TWO pools for it, and they are not interchangeable: a **110-row pool** ranking every dive (revisit dives included, so BIPB70B/72B/69C compete) and a **99-site pool** ranking one value per site. The rank quoted below is on the 110-row dive pool. For this site the canonical value is **24.21%, rank 8 of 110** (1st training: 24.21%, rank 6 of 110). The two runs agree to <0.05 pp on 100 of 110 rows and diverge only on the bleaching-onset dives, so for most sites the distinction changes the rank but not the number.

**Benthic data:** [Download BICC50 benthic survey data (CSV)](../benthic_data/BICC50_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).