### BIPB112 — Peros Banhos
**CRD page:** [Peros Banhos Coral Survey Sites (Apr 30–May 3, 2015)](https://coralreefdiagnostics.com/peros-banhos-20150430-to-0503)

**02 May 2015**
**GPS:** 5.23730°S, 71.96200°E
**Setting:** exposed (windward) emergent fringing forereef, 32.7 m
**Coral cover:** 45.0% (16 formal transects across 4 depth strata, D2–D5; **algal cover 40.7%**) · **published image bleaching:** 15.9%
**Live hard coral cover across methods:** formal transects 45.0% · CoralNet points 29.1% · Coralscapes masks 27.3% (15 frames). These products have different sampling footprints and denominators and should not be averaged.
**Temperature:** 29.8 °C · **Salinity:** 34.4
**Photo review:** all 15 of 15 jpg stills examined (single dive, 02 May 2015, 14:57–15:38 UTC by file `DateTimeOriginal`; the folder also holds one 32-byte video stub — see Flags)
**Representative photos:** P5021018.jpg, P5021010.jpg, P5020986.jpg, P5021028.jpg, P5020980.jpg

BIPB112 is a windward, emergent fringing forereef on the outside of Peros Banhos at 32.7 m, and it carries **above-average coral cover for the archipelago: 45.0% coral against 40.7% algae, ranking 36th of 109 Chagos sites for coral cover and mid-pack (54th of 109) for algal cover.** The substrate is hard and largely clean: bare substrate (PB) is only 3.1%, and turf-plus-sediment (PTSTS) an almost negligible 0.66% — sediment simply does not accumulate on this exposed slope. What algae there is splits between crustose coralline algae (PCCATS) at 18.6% of total benthos (32nd of 109) and fleshy turf (PTTS) at 12.4% (37th of 109), with macroalgae (PMTS) 6.6% and erect algae (PETS) 2.3%. **This is also one of the few Chagos sites with a nonzero cyanobacteria reading — PCYTS 0.04%**, effectively a trace but explicitly recorded rather than blank. Non-coral invertebrates (PITS) hold 11.2%. Coral, algae, invertebrates and bare rock sum to exactly 100.0% at the site level.

Genus richness is high: **30 nonzero genera, 16th of 109 Chagos sites**, and the 30 genus percentages sum to 45.0417% 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 20.21% of the benthos — **45% of all the coral at the site** — followed by *Porites* 10.78% (24%), *Montipora* 5.65% (13%), *Pocillopora* 2.43% (5%) and *Astreopora* 1.13% (3%); those five hold 89% of the coral recorded. *Seriatopora* (1.10%), *Goniastrea* (0.59%), *Pavona* (0.48%), *Pachyseris* (0.35%), *Cyphastrea* (0.31%) and *Leptastrea* (0.31%) follow, then a long tail of 19 further genera — *Platygyra*, *Symphyllia*, *Isopora*, *Favia*, *Favites*, *Leptoseris*, *Psammocora*, *Montastrea*, *Cantharellus*, *Acanthastrea*, *Ctenactis*, *Millepora*, *Oxypora*, *Turbinaria*, *Fungia*, *Lobophyllia*, *Stylophora*, *Echinopora* and *Galaxea* — each under 0.26% of the benthos. The tail is what earns the site its 16th-of-109 richness placement: it holds only about 2.4 points of benthos in total, but it holds them across nineteen genera.

**The four depth strata contain the real finding at this site: an almost complete handover of the reef from *Acropora* to *Porites* and *Montipora* between the shallowest and deepest station.** D2 (6 transects, shallowest sampled) carries the profile's highest coral cover, 56.8%, against 36.5% algae — and it is overwhelmingly monogeneric: ***Acropora* alone reaches 42.47% of the benthos, 75% of all the coral on that transect***, with *Porites* 10.17% a distant second and *Montipora* (1.10%), *Pocillopora* (1.03%), *Pavona* (0.47%) and *Platygyra* (0.37%) barely registering. At D3 (6 transects) cover falls to 38.6% and algae rises to the profile's peak of 54.5% (CCA 26.0%, turf 22.8%); *Acropora* drops by well over half to 17.03% (44% of that transect's coral), while *Porites* holds at 9.23% and *Montipora* climbs to 4.43%. D4 (3 transects) is similar at 39.8% cover and 41.7% algae, but the order shifts again — *Acropora* 15.33%, then ***Montipora* overtakes *Porites*** at 7.07% against 6.73%, with *Pocillopora* 2.53%, *Goniastrea* 1.73% and *Pachyseris* 1.40%. **At D5 the reversal is complete: cover recovers to 45.0% but *Acropora* collapses to 6.00% — just 13% of the coral present — while *Porites* leads at 17.00% (38%) and *Montipora* reaches 10.00% (22%), with *Astreopora* and *Seriatopora* both at 4.00% and *Pocillopora* 3.00%.** D5 also carries the profile's lowest algal cover (30.0%) but its highest macroalgal fraction (PMTS 13.0%, against 1.7–7.3% shallower) and zero turf-plus-sediment. **The site-level figures of 20.21% *Acropora* and 10.78% *Porites* therefore describe no single transect**: they average a shallow station where *Acropora* holds three-quarters of the coral against a deepest station where it is a minor taxon and *Porites* plus *Montipora* hold three-fifths of it. **D5 rests on a single transect** and must be read as indicative only (see Flags).

The photographs — all 15 stills — document a tabular-*Acropora* forereef of considerable structural scale, and they line up closely with the depth story above. The signature architecture is large, tiered tabular colonies: P5021018 shows a single table well over a metre across, P5021008 three overlapping tables on a ledge, P5021017 and P5021025 broad tabular fields extending into blue water, and P5021026 a wide shot in which two dozen plate-and-table colonies are visible across the slope at once. P5021027 records the other structural mode — a dense, near-continuous field of small overlapping plates consistent with *Montipora* and plating *Acropora* over pavement, the visual counterpart of the *Montipora* increase at D3–D4. P5021024 is a close-range frame of a mixed thicket where pale lavender and cream digitate/corymbose colonies consistent with *Acropora* and knobby clumps consistent with *Pocillopora* sit over a pink coralline-crusted surface beside a large tan plate. P5020978 looks along the drop-off itself, showing a steep, structurally complex slope with plates, gorgonians and wire corals falling into deep blue water — consistent with the 32.7 m survey depth and the windward setting. P5020966 is the only close substrate frame and it is instructive: a dense mat of pale calcified branching macroalgae (*Halimeda*-type) over maroon coralline crust, with encrusting sponges and a small green brittle star — a good illustration of where the site's 6.6% macroalgal cover actually sits. Water clarity is excellent throughout.

**Condition is a prominent finding in this set, and the imagery is consistent with the site's top-quartile published figure.** The published image-based bleaching value is **15.9%, ranking 13th of 98 Chagos sites carrying a value — well inside the most-bleached quartile of the archipelago.** Bleaching in the frames is unambiguous and colony-scale rather than marginal. **P5021018 is the single clearest record: a large tiered tabular colony rendered uniformly bone-white across its entire upper surface, with fully pigmented tan colonies in the same frame.** P5020986 shows a cluster of four or more white tables on the upper slope beside a fully white encrusting/foliose colony. P5021010 is a close view of a completely white plate colony with a partially bleached, tan-mottled neighbour to its left — the two together showing the intermediate and complete stages side by side. P5021025 and P5021026 both show white and bleach-mottled tables scattered through otherwise pigmented fields, and P5021008 shows a large pale-cream table adjoining healthy tan ones. Because a single still cannot separate a freshly bleached colony from a very recently dead one, these are recorded as **frame-level bleaching-or-recent-mortality observations, not as a recalculated site percentage**. No disease lesion or tissue-loss margin, no *Drupella* aggregation, no crown-of-thorns starfish and no predation scarring was identified in any of the 15 frames.

Fauna is sparse in the benthic frames — scattered damselfish, anthias and butterflyfish — but the set carries one clear pelagic record: **P5020980 shows a loose school of at least five bluefin-trevally-type jacks (*Caranx*-type, hedged from a single frame) cruising in open blue water off the drop-off.** **The photo folder is named `BiPb112-whale shark!-finished`, but no whale shark appears in any of the 15 stills.** The only other file in the folder is `P5020960.MOV`, and it is a 32-byte stub containing nothing but a QuickTime `ftyp` header — the content is unavailable, so if the whale shark encounter was recorded, it was recorded there and cannot be verified from this review. This is flagged rather than asserted; the folder name is not by itself a sighting record. P5021028 shows two divers working a transect tape across the reef — direct documentation of the formal survey that produced the cover figures above.

<b>Assemblage composition.</b> Formal transects give 45.0% coral against 40.7% algae across 30 nonzero genera, led by *Acropora* (20.2% of benthos, 45% of all coral), *Porites* (10.8%), *Montipora* (5.7%), *Pocillopora* (2.4%) and *Astreopora* (1.1%), with a nineteen-genus tail holding only ~2.4 points of benthos between them. The depth profile is a genus handover: 56.8% cover at D2 where *Acropora* alone holds 75% of the coral, through 38.6% and 39.8% at D3 and D4 where *Montipora* rises, to 45.0% at D5 where *Acropora* is down to 13% of the coral and *Porites* (38%) plus *Montipora* (22%) lead. The photographs match — metre-scale tiered tables, dense plating fields, a gorgonian-hung drop-off and *Halimeda*-type macroalgal mats — together with clear, colony-scale bleaching in at least five frames and a school of trevally-type jacks off the wall.

**Flags:** (1) **BLEACHING SOURCE CONFLICT — three different values exist for this site and someone needs to adjudicate them.** The published `Mean(%bleaching)` column of `Chagos benthic data by site (n=109).csv` gives **15.92%**, and that is the value quoted in the header and used for the 13th-of-98 rank. But `CHAGOS_SITE_REFERENCE_20260825.csv` carries `pct_bleaching_published` = **38.22%** for BIPB112, and `chagos_pct_bleached_recomputed.csv` recomputes **12.67%** (19 bleached of 150 coral points, 450 points, 15 images) against a `pct_bleached_in_master_md` of **38.2%**, a **−25.53 pp delta explicitly marked `MISMATCH`**. Three sources, three numbers, spanning 12.7–38.2%. This shard follows the standing rule and quotes the published site-table column verbatim; the discrepancy is unresolved and is the most actionable item at this site. (2) **Multi-method deviation is substantial but smaller than at most Chagos sites.** CoralNet points (29.1%) and Coralscapes masks (27.3%) sit 16–18 percentage points below the formal-transect figure (45.0%; deviation range −17.7 to −15.9 pp). Do not average the three figures; the transect value is treated as ground truth for cover. (3) **The CV figures rest on a 15-frame scored subset**, which happens to equal the jpg count in the folder — that coincidence does not make the products comparable, and the denominators still differ. (4) **The CoralNet points figure is a CV product, not verified ground truth**; point-annotation data corpus-wide is roughly 76% unconfirmed machine output. (5) **D5 rests on ONE transect** (against 6, 6 and 3 at D2, D3 and D4), so the deepest station's striking *Porites*/*Montipora* dominance, its 45.0% cover and its 13.0% macroalgal reading are all single-transect values and are indicative rather than replicated. The D2→D5 handover is nonetheless supported by the monotonic *Acropora* decline across all four strata (42.47 → 17.03 → 15.33 → 6.00), which does not depend on D5 alone. (6) **Depth codes D2–D5 carry no metre boundaries** in `Chagos benthic data by site-depth (n=336 rows).csv`; they are ordered shallow-to-deep strata, and the 32.7 m in the header is the site-level survey depth, not the depth of any one transect. **No D1 stratum exists for this site**, so the shallowest reef zone is unsampled. (7) **`P5020960.MOV` is a 32-byte stub** — a QuickTime `ftyp` header and nothing else, with a file creation date of 2019 rather than 2015. The content is unavailable locally (evicted iCloud placeholder or truncated file) and was not reviewed. **Given the folder name, this stub may be the whale shark record**; it should be restored from the original media before the site's megafauna history is treated as documented. (8) **The folder name `BiPb112-whale shark!-finished` asserts a whale shark encounter that NO reviewed still supports.** All 15 jpgs were examined; the only pelagic record is the trevally-type school in P5020980. The folder name is reported here as an unverified claim, not as an observation. (9) **Exposure is recorded twice and consistently** — `exposure-original` = windward, `exposure-revised` = exposed — so no conflict at this site, unlike some Chagos records. (10) **No COTS, disease, *Drupella* or predation scarring was observed** across the full 15-frame review — an explicit absence-of-evidence finding. (11) **White colonies cannot be separated from recently dead ones in a single still**; the bleaching records above are "bleached or recently dead", and no site percentage is derived from them. (12) **Two divers and a transect tape are visible in P5021028** — standard field documentation of the survey, not a data concern. (13) **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. (14) **Ranks off the `Mean(%bleaching)` column are approximate** — that column carries at least one known bad value (BITB58 listed at 22.22% against a corrected 0.77%), so the 13th-of-98 placement has not been independently audited, and flag 1 makes it doubly provisional here. (15) **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 **38.22%, rank 1 of 110** (1st training: 15.92%, rank 17 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 BIPB112 benthic survey data (CSV)](../benthic_data/BIPB112_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).