### BIPB60 — Peros Banhos
**CRD page:** [Peros Banhos Coral Surveys (Apr–May 2015)](https://coralreefdiagnostics.com/peros-banhos-)

**14 Apr 2015**
**GPS:** 5.44336°S, 71.74987°E
**Setting:** intermediate-exposure (published as windward) emergent fringing forereef, outside the lagoon, 32.0 m
**Coral cover:** 42.2% (19 transects across 4 depth strata, D1–D4; **algal cover 39.5%**) · **published image bleaching:** 11.8% (`Mean(%bleaching)` column)
**Live hard coral cover across methods:** formal transects 42.2% · CoralNet points 26.2% · Coralscapes masks 27.8% (n_photos = 36). These products have different sampling footprints and denominators and should not be averaged.
**Temperature:** 30.0 °C · **Salinity:** 34.3
**Photo review:** 25 of 44 frames examined — all 6 top-level frames, plus a systematic every-other sample of 19 of the 38 benthic frames held in the `BiPb60-CoralNet` subfolder, spanning P4140559 through P4140626
**Representative photos:** P4140564.jpg, P4140575.jpg, P4140606.jpg, P4140604.jpg, P4140585.jpg

**Correction, 2026-08-26 — this site is NOT photographically undocumented.** The previous version of this shard stated that "all 6 frames" had been examined and that "this is not a benthic photo set," on the basis of the six frames visible at the top level of `BiPb60-finished`. **That folder also contains a `BiPb60-CoralNet` subfolder holding 38 further frames (P4140559–P4140627), and they are a proper benthic set** — wide-angle reef panoramas, coral-garden close-ups, condition frames and fauna. Everything the earlier reviewer recorded from the six top-level frames is preserved below; the reef description that the old shard said could not be written now can be.

BIPB60 is a middling-cover, algae-heavy fringing forereef on the outer rim of Peros Banhos. **Coral cover of 42.2% ranks 43rd of 109 Chagos sites; algal cover of 39.5% ranks 56th of 109** — this is a reef where coral and algae are close to parity, and the balance is the site's defining number. The algal budget is dominated by crustose coralline algae, which alone hold **21.2% of the benthos, 14th-highest of 109 sites** — one of the most heavily CCA-encrusted reefs in the archipelago. Fleshy turf follows at 11.6% (42nd of 109), then erect algae 3.6%, sediment-laden turf 1.7%, macroalgae 1.3% and cyanobacteria 0.1%. Bare substrate is low at 9.2%, and non-coral invertebrates are unusually high at 9.0%. Coral, algae, bare substrate and invertebrates sum to exactly 100.0% of the benthos, so nothing is unallocated. The reading is a hard, well-grazed, coralline-encrusted reef rather than a weedy one: "39.5% algae" here means mostly pink encrusting pavement, not macroalgal growth, and the photographs confirm it — pink and mauve CCA coats the rock in almost every benthic frame.

The assemblage is genus-diverse and emphatically *Porites*-built. **24 non-zero genera were recorded, 60th of 109 Chagos sites** — mid-pack for the archipelago, and substantially more than the 18 quoted in the earlier version of this shard. All values below are **percent of total benthos, not percent of coral**, and the 24 genus percentages sum to 42.23%, an exact match to the published cover figure. ***Porites* holds 22.83% of the benthos — 54% of every coral point recorded at the site** — with *Acropora* a distant second at 8.39% (20%), then *Montipora* 3.14% and *Pocillopora* 3.12%. Below that is a twenty-genus tail, every member under 0.7%: *Millepora* 0.68%, *Heliopora* 0.48%, *Stylophora* 0.47%, *Pavona* 0.46%, *Leptastrea* 0.46%, *Goniastrea* 0.46%, *Montastrea* 0.44%, *Astreopora* 0.43%, *Goniopora* 0.25%, *Favia* 0.25%, *Favites* 0.14%, *Galaxea* 0.08%, *Cyphastrea* 0.04%, *Gardineroseris* 0.04%, *Cantharellus* 0.03%, *Turbinaria* 0.03%, *Coscinaraea* 0.01%, *Fungia* 0.01%, *Platygyra* 0.01% and *Leptoseris* 0.01%.

**The depth structure is a clean genus handover, and it is the real finding at this site.** Coral cover peaks at D2 (49.3%) and then declines steadily with depth — D3 39.1%, D4 36.3% — while the shallowest stratum, D1, holds 44.3%. The compositional shift across that peak is what matters. **At D1, *Acropora* leads at 19.00% of the benthos against *Porites* at 12.80%**, with *Montipora* 4.40%, *Millepora* 2.70%, *Stylophora* 1.80% and *Pocillopora* 1.70% behind them, across 11 genera. **From D2 down, the order inverts and never recovers**: *Porites* takes **33.03%** at D2 against *Acropora*'s 6.47% (12 genera), then **25.63% versus 4.49%** at D3 (14 genera), then **19.85% versus 3.60%** at D4 (19 genera). The site-level figure of 22.83% *Porites* is therefore not a uniform value but the average of a reef that is *Acropora*-fronted in the shallows and *Porites*-built below them. Two further gradients run alongside it. Algal cover is highest in the shallows and lowest at depth (47.5% at D1, 40.0% at D2, 41.9% at D3, 28.7% at D4), so the shallow band is the algae-heaviest part of the reef. And **genus richness rises as cover falls** — 11 genera at D1 against 19 at D4, where *Montastrea*, *Goniopora*, *Galaxea*, *Turbinaria*, *Coscinaraea*, *Fungia* and *Platygyra* all appear and the assemblage is at its most even. The deepest stratum also carries the site's highest non-coral invertebrate share (18.7% at D4 against 4.5–6.9% shallower).

**Unlike some Chagos sites, the headline number here is robust to weighting.** Site cover is the unweighted mean of the four per-stratum means, 42.23%; re-computed as a transect-weighted mean over the same 19 transects (D1 n = 2, D2 n = 6, D3 n = 7, D4 n = 4) it gives 42.26% — a 0.03 pp difference. The one thin stratum is **D1, resting on two transects**, and it is the stratum carrying the site's only *Acropora*-led composition, so the shallow half of the depth story above is less replicated than the deep half.

The benthic frames show a structurally complex, high-diversity reef of moderate relief. Wide-angle panoramas (P4140602, P4140608, P4140612) document a coral garden of overlapping tabular and corymbose *Acropora*, olive and tan massive colonies consistent with *Porites*, brown digitate and knobby colonies consistent with *Pocillopora* and *Stylophora*, and encrusting growth over pink coralline pavement, with coral patches separated by rock rather than sand. **P4140604** is the clearest structural frame: three or four *Acropora* tables stacked at different heights above a shaded understorey, with a mauve-purple *Pocillopora*-type colony and yellow-green corymbose colonies between them. **P4140612** shows the lobate, orange-brown columnar growth form that dominates the *Porites* signal in the transect data. P4140590 records a high-relief bommie carrying branching yellow *Millepora*-type fire coral — the transects put *Millepora* at 0.68% of the benthos, and it is visibly present. Diver activity and the survey itself appear in **P4140559**, where a diver works in the background above open pavement scattered with dead, sediment-veneered tabular plates. Water clarity is good throughout, with strong blue backscatter and pink CCA rendering cleanly.

**Condition is the weakest element at this site, and for the first time the photographs bear on it directly.** The published image-based bleaching figure is **11.8%, which ranks 21st of 98 Chagos sites with a value — the top quartile of bleaching severity** in the archipelago and an order of magnitude above the low-bleaching majority. The benthic frames corroborate that. **P4140564 is the strongest single record: a large complex of bone-white, ridged encrusting/massive colonies covering most of the frame's mid-ground**, with normally pigmented olive and grey colonies immediately adjacent under the same lighting, and dark surgeonfish grazing across it. **P4140575** shows a *Pocillopora*-type colony in which one branch cluster at lower right is entirely white while the rest of the colony is normally pigmented brown-olive, and a second colony above it carries white branch tips well beyond the extent of normal growth margins — classic partial bleaching. **P4140600** contains a small, pure-white ridged massive colony sitting on coralline pavement at lower left, isolated among fully pigmented neighbours. Several *Acropora* tables in P4140566 read as unusually pale grey-white, but plate *Acropora* is naturally light-toned and the frame is shot from above, so those are noted rather than scored. Taken together the imagery supports the published figure as real rather than an artefact — this reef was carrying visible bleaching on 14 Apr 2015. **No disease band, tissue-sloughing front, *Drupella* aggregation, parrotfish scarring or crown-of-thorns starfish was identified in any of the 25 frames reviewed.**

Fauna is abundant and includes two large-animal encounters. **A tawny nurse shark (*Nebrius ferrugineus*) is photographed at close range in P4140606** — uniformly brown, blunt broad head, small eye, long upper caudal lobe — cruising immediately above corymbose *Acropora* and massive *Porites*. **A second shark appears in the mid-field of P4140610**, over open reef; the profile is consistent with a requiem shark or the same nurse shark, and the frame does not resolve it further. Reef fish are numerous: blue-green parrotfish, yellow butterflyfish and clouds of anthias in P4140590; Moorish idols, sergeant majors, a wrasse and an anemonefish with its host anemone in P4140566; damselfish and butterflyfish throughout. The two backlit top-level frames (P4140562, P4140563) record a substantial school of unicornfish or surgeonfish silhouetted against the surface light, and **also show gorgonian sea fans and thin black whip or wire corals standing off the substrate — neither of which appears in the transect genus list**; P4140563 contains one elongate fish whose silhouette does not permit a call between a small shark and a barracuda. The remaining four top-level frames (P4140585–588) are a macro sequence of a **single phyllidiid nudibranch** — black ground, pale blue-white ridges, yellow-orange tipped tubercles — photographed from four angles on coralline-crusted pavement; *Phyllidia*-type, with species not resolvable from these images alone.

<b>Assemblage composition.</b> Formal transects give 42.2% coral against 39.5% algae (itself 21.2% crustose coralline, the 14th-highest CCA share of 109 sites), 9.2% bare substrate and 9.0% non-coral invertebrates, across 24 non-zero genera of which ***Porites* alone is 54%**. The four-stratum depth profile shows a clean inversion: *Acropora*-led at D1 (19.00% vs 12.80%), *Porites*-led at every stratum below it (33.03%, 25.63%, 19.85%), with genus richness rising and cover falling toward D4. The 25 frames reviewed document that reef directly — stacked *Acropora* tables, lobate *Porites*, fire coral, extensive pink coralline pavement, a tawny nurse shark, a second shark, gorgonians and whip corals absent from the transect list, a phyllidiid nudibranch, and **three frames carrying unambiguous colony-scale bleaching**, consistent with a published bleaching figure in the archipelago's worst quartile.

**Flags:** (1) **The previous shard's central claim was wrong and has been corrected.** It reported "all 6 frames examined — this is not a benthic photo set" and recommended treating BIPB60 as photographically undocumented. The site in fact holds **44 frames, 38 of them in the `BiPb60-CoralNet` subfolder**, and they are a full benthic set. Any downstream text, popup or table repeating "undocumented" for BIPB60 needs the same correction. **Note also that `sitefacts.py` does not recurse into subfolders** — it reports `frames = 7` for this site, counting the subfolder name itself as a frame stem. That is a tooling bug likely to have produced the same error at other sites with `-CoralNet` subfolders. (2) **Frame-count mismatch across three sources:** 44 frames in the folder tree, 38 in the CoralNet subfolder, and `n_photos = 36` in `chagos_multimethod_cover_by_site.csv`. The CoralNet and Coralscapes figures rest on that 36-frame scored subset, not on 38 and certainly not on 44; the two frames' worth of difference is unexplained. Never print the folder count beside a CV figure. (3) **The earlier shard's genus count was wrong: 24 non-zero genera, not 18.** Its cover rank ("39th of 98") and bleaching rank ("31st of 109") were also not from the central rank file; the correct values are **cover 43rd of 109, bleaching 21st of 98** on the `Mean(%bleaching)` column. Its depth-band metre labels ("8–12 m", "12–18 m", "18–25 m") had no source in the published tables and have been replaced with the D1–D4 stratum codes; its per-stratum *Acropora* values at D3 and D4 (5.2%, 4.8%) were also slightly off the published 4.49% and 3.60%. (4) **Bleaching at 11.8% is high for Chagos (21st of 98) and is now photographically corroborated** — P4140564 (large white encrusting/massive complex), P4140575 (partially bleached *Pocillopora*-type colony) and P4140600 (small white massive colony) each show colony-scale whitening beside normally pigmented neighbours under identical lighting. This is a real bleaching observation to carry into the running COTS/bleaching/disease log, not a model artefact. (5) **The multi-method deviation is −16.0 to −14.4 pp** (42.2% transect vs 26.2% CoralNet vs 27.8% Coralscapes) — moderate by Chagos standards, but the three figures still must not be averaged; the transect value is treated as ground truth for cover. (6) **The CoralNet points figure is a CV product, not verified ground truth.** Point-annotation data corpus-wide is roughly 76% unconfirmed machine output. (7) **Gorgonians and black whip/wire corals are visible in P4140562–563 but absent from the transect list**, which records scleractinians plus *Millepora* and *Heliopora* only — the octocoral and antipatharian component of this reef is unquantified. (8) **D1 rests on only two transects**, and it is the single stratum where *Acropora* outranks *Porites*; the shallow half of the depth inversion is the less replicated half. The site figure is otherwise robust to weighting (unweighted 42.23% vs transect-weighted 42.26%). (9) **The 22.8% site-level *Porites* figure averages a genuine depth inversion** and should never be quoted without the depth breakdown. (10) **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`). Note that the `Mean(%bleaching)` column is **not** a mean — it equals the first CoralNet training run, while the dive-history map publishes the second. At BIPB60 the two runs agree (11.76% and 11.759%, ranking 27th and 31st of 110 respectively), so the divergence problem does not bite here; but the column carries at least one known bad value elsewhere (BITB58 at 22.22% where both runs give ~0.77%), so any rank taken off it is approximate. (11) **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 **11.76%, rank 31 of 110** (1st training: 11.76%, rank 27 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 BIPB60 benthic survey data (CSV)](../benthic_data/BIPB60_benthic.csv) — scleractinian cover by genus with full names, plus substrate and algal categories (bare substrate, CCA, turf, sediment-laden turf, macroalgae, erect algae, cyanobacteria, non-coral invertebrates), from the LOF BIOT-1 survey (Mayfield & Dempsey 2026, *Oceans*, supplemental file S1).