### BISA107 — Salomon
**CRD page:** [Salomon Islands Coral Survey Sites (Apr 26–30, 2015)](https://coralreefdiagnostics.com/salomon-20150426-to-0430)

**30 Apr 2015**
**GPS:** 5.31024°S, 72.26888°E
**Setting:** exposed emergent forereef outside the lagoon, 34.75 m (the `exposure-original` cell is blank — see Flags)
**Coral cover:** 24.6% (3 depth transects, D2–D4; **algal cover 65.4%**) · **published image bleaching:** 7.1%
**Live hard coral cover across methods:** formal transects 24.6% · CoralNet points 12.5% · Coralscapes masks 12.8% (12 frames). These products have different sampling footprints and denominators and should not be averaged.
**Temperature:** 29.3 °C · **Salinity:** 34.4
**Photo review:** 20 of 39 jpg stills examined — a systematic every-second-frame sample spanning the whole dive sequence from P4300837 to P4300876 (P4300837, 839, 841, 843, 846, 848, 850, 852, 854, 856, 858, 860, 862, 864, 866, 868, 870, 872, 874, 876). **P4300839 is a monochrome/desaturated frame and was excluded from colour-based condition scoring** (see Flags). This folder is labelled a *fun dive* and several sampled frames carry no benthos at all.
**Representative photos:** P4300856.jpg, P4300860.jpg, P4300846.jpg, P4300852.jpg, P4300870.jpg

BISA107 is a deep exposed forereef on the outside of Salomon atoll, surveyed at 34.75 m, and **it is the most algae-dominated site in this batch and one of the most algae-dominated in the whole archipelago: 24.6% coral against 65.4% algae, ranking 85th of 109 Chagos sites for coral cover and 2nd of 109 for algal cover.** The composition of that algal budget is what makes the site genuinely exceptional rather than merely degraded. **Crustose coralline algae (PCCATS) alone accounts for 34.9% of total benthos — the single highest CCA value of any of the 109 Chagos sites, 1st of 109.** Macroalgae (PMTS) adds 12.2%, encrusting algae (PETS) 7.5%, fleshy turf (PTTS) only 8.6% (65th of 109), turf-and-sediment 2.2% and cyanobacteria 0.03%; those six categories sum to 65.433%, an exact match to the published algal cover. Bare substrate is 7.3% and non-coral invertebrates only 2.6% — the lowest invertebrate value in this batch. **A reef where more than a third of the entire benthos is coralline crust is not an algal-overgrowth story: it is a hard, well-grazed, wave-swept forereef whose dominant surface is cemented pink pavement rather than weed**, and the photographs show exactly that.

Coral cover is spread across **25 nonzero genera, 56th of 109 Chagos sites for genus richness — mid-pack** — and the 25 genus percentages sum to 24.6000% 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 10.10% — **41% of all the coral present** — followed by *Stylophora* at 3.67% (15%), an unusually high placing for that genus anywhere in this dataset. Then *Montipora* 1.40% (6%), *Pocillopora* 1.23% (5%), *Favia* 1.17% (5%), *Isopora* 1.10% (4%) and *Favites* 1.00% (4%). **The striking absence is *Porites*, which at 0.87% ranks only 8th here and holds a bare 4% of the coral** — at most Chagos forereefs it is first or second. *Millepora* (0.67%), *Platygyra* (0.63%), *Gardineroseris* (0.53%) and *Montastrea* (0.43%) follow, then a tail of thirteen genera each below 0.35%: *Heliopora*, *Pavona*, *Fungia*, *Cantharellus*, *Cyphastrea*, *Echinopora*, *Leptoseris*, *Leptastrea*, *Pachyseris*, *Psammocora*, *Goniastrea*, *Lobophyllia* and *Turbinaria*.

**The three depth transects show coral cover more than doubling from the shallowest to the deepest station while algal cover falls — yet algae outweighs coral at every single depth, including the best one.** D2, the shallowest station sampled, carries just 16.4% coral against **70.7% algae**, with 9.1% bare substrate and 3.8% non-coral invertebrates; the assemblage (16 genera) is led by *Acropora* 6.10%, *Stylophora* 2.30% and *Millepora* 2.00% — **the only transect where *Millepora* occurs at all** — with *Pocillopora* 1.30%, *Montipora* 1.00%, *Porites* 0.80%, *Favia* 0.70% and *Heliopora* 0.50%. At D3 cover rises modestly to 20.4% against 68.5% algae, and **the lead changes hands: *Stylophora* takes over at 6.30% of the benthos — 31% of that transect's coral and nearly twice its own site-level value** — ahead of *Acropora* 3.70%, *Montipora* 2.20%, *Favia* 1.40% and *Gardineroseris*, *Montastrea* and *Pocillopora* at 1.10% each (18 genera). At D4, the deepest station, **coral cover nearly doubles again to 37.0% while algae falls to 57.1% and bare substrate drops to 3.8%; *Acropora* surges to 20.50% — 55% of that transect's coral, twice its site-level value** — with *Isopora* 3.10% (a genus effectively confined to this depth), *Stylophora* 2.40%, *Favites* 2.30%, *Platygyra* 1.50%, *Favia* 1.40%, *Pocillopora* 1.30% and *Porites* 1.30%, on 20 genera. All three depth rows account for 100% of the benthos with no shortfall. **The site-level figures describe no single transect**: 10.10% *Acropora* averages a shallow station at 6.10% against a deep station at 20.50%, and 3.67% *Stylophora* averages two modest values against a mid-depth peak of 6.30%.

The photographs corroborate the coralline-pavement finding directly and forcefully. **P4300856 and P4300858 are the defining frames: a broad, gently sloping reef surface running to the limit of visibility, its entire ground a pale pink-to-mauve cemented crust with only small, widely spaced coral colonies on top** — the visual signature of a site carrying the archipelago's highest CCA cover. P4300854, P4300866, P4300868, P4300870 and P4300874 show the same low-relief, low-cover, crust-dominated habitat with scattered plating, encrusting and small branching colonies. P4300860 and P4300862 document the other face of the site — **a steep wall section carrying large olive-and-gold gorgonian sea fans standing off into blue water**, with dense anthias clouds against the drop. P4300850 shows blue-and-yellow fusiliers over pavement carrying yellow-green plating colonies. P4300864 and P4300872 show the reef crest with dense small-fish aggregations above it. Image-based genus calls are tentative throughout and no cover figure is derived from them.

**Two frames record things that are not benthos and matter more than the benthos.** **P4300846 documents an abandoned fishing net draped across the reef — an intact panel of dark netting with visible mesh, lying over coralline pavement, plating corals and small branching colonies.** This is marine debris / derelict fishing gear on a reef inside the BIOT no-take MPA, and it is the single most actionable observation in this batch. **P4300852 records a spiny lobster (*Panulirus*-type) sheltering under a ledge**, antennae extended, beneath grey-white plating colonies and pink coralline crust.

**Condition reads as unremarkable and the imagery does not contradict it.** The published image-based bleaching figure is **7.1%, ranking 50th of 98 Chagos sites carrying a value — squarely at the archipelago median.** A handful of small white or very pale colonies appear scattered through the wide-field frames (P4300850, P4300854, P4300866), but at the working distance of those shots they cannot be separated from naturally pale colonies, freshly exposed skeleton or coralline crust, and **no colony in this set shows the unambiguous close-range white-against-pigmented contrast that would justify a bleaching call.** Colour interpretation is additionally complicated here because a CCA-dominated surface is itself pale pink to near-white and covers more than a third of the benthos — the exact condition under which a colour-based annotator is most likely to mis-assign pallor. **No disease lesions, tissue-loss bands, *Drupella* aggregations, crown-of-thorns starfish or predation scarring were seen in any of the 20 frames examined.**

Fauna is conspicuous, and sharks are the recurring element. **Three separate shark encounters are documented in the sampled frames — P4300837 (an individual over the reef, seen from above), P4300841 (in open blue water off the slope) and P4300843 (cruising low over the reef among damselfishes)** — and a fourth shark appears in mid-water beside a diver in the monochrome frame P4300839. Dense anthias and fusilier clouds hang over the crest and the wall throughout, with parrotfish, moorish idols, groupers and damselfishes in the closer frames. **Divers appear in a substantial share of the sampled frames** (P4300839, P4300850, P4300866, P4300868, P4300872, P4300874, and P4300876, which is entirely a group of divers at the surface with no benthos in view) — consistent with the folder's *fun dive* label and relevant to how this photo set should and should not be used.

<b>Assemblage composition.</b> Formal transects give 24.6% coral against 65.4% algae across 25 nonzero genera, but the algal figure is dominated by crustose coralline algae at 34.9% of total benthos — **the highest CCA value of any of the 109 Chagos sites** — with only 8.6% fleshy turf. *Acropora* holds 41% of the coral, *Stylophora* an unusually high 15%, and *Porites* an unusually low 4%. The depth profile runs 16.4% coral at D2, 20.4% at D3 and 37.0% at D4, with *Millepora* confined to the shallowest transect, *Stylophora* peaking at the middle one, and *Acropora* and *Isopora* surging at the deepest — yet algae outweighs coral at every depth. The photographs corroborate a low-relief, coralline-crusted forereef with a gorgonian-hung wall section, plus three shark encounters, a spiny lobster, and an abandoned fishing net lying on the reef.

**Flags:** (1) **AN ABANDONED FISHING NET IS PHOTOGRAPHED ON THE REEF AT THIS SITE (P4300846)** — an intact panel of dark mesh netting draped over coralline pavement and living coral. This is derelict fishing gear inside the BIOT MPA and should be carried into any marine-debris, IUU-fishing or reef-impact record, not left buried in a site description. Frame P4300846 is the evidence. (2) **This folder is labelled a "fun dive"** (`BiSa107-fun dive-finished`) and the imagery reflects it: the set is a recreational photographic record, not a systematic benthic photo-transect. Several sampled frames carry no benthos at all (P4300876 is entirely divers at the surface; P4300837, P4300839 and P4300841 are water-column shark shots), and divers are visible in at least seven of the 20 frames reviewed. **The photographs must not be treated as a representative sample of the substrate**, and any cover figure derived from them will be biased by frame selection. Note that formal transect data DO exist for the site (three depth transects, D2–D4) — the *fun dive* label describes the photography, not the survey. (3) **Severe frame-count mismatch:** the photo folder holds 39 jpg stills but `chagos_multimethod_cover_by_site.csv` records **`n_photos = 12`** for the CV comparison — under a third of the folder, and by far the largest such gap in this batch. The most likely explanation is that only the benthic-facing frames were scoreable, which is consistent with flag (2), but this was not verified. **The CoralNet (12.5%) and Coralscapes (12.8%) figures rest on that 12-frame subset, and 12 frames selected from a recreational set is a thin and possibly non-random basis for a site-level cover estimate.** Never print the 39-frame folder count beside those figures. (4) **P4300839 is a monochrome / fully desaturated frame** (a diver and a shark in mid-water). It must be excluded from colour-based condition scoring and from any CV cover annotation, in the same way as the known fluorescence frames (BISA104 P4290914–915, BISA103 P4290870, and BISA102 P4280877 flagged in this batch). Whether P4300839 falls inside the 12-frame scored subset was not verified and should be checked. (5) **This site holds the highest crustose coralline algal cover in the Chagos dataset: PCCATS 34.9%, 1st of 109**, and the second-highest total algal cover, 65.4%, 2nd of 109. Both are notable enough to be quoted directly rather than folded into an "algae-dominated" summary — the site is coralline-pavement dominated, not weed-dominated (fleshy turf is only 8.6%, 65th of 109). (6) **Bleaching interpretation is unusually unreliable at this site and the published figure deserves extra scepticism.** A CCA-dominated surface is naturally pale pink to near-white and covers more than a third of the benthos, which is precisely the condition under which a colour-based annotator is most likely to mis-assign pallor. No bleaching call was made from these photographs. (7) **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. Any rank taken off that column is approximate: it carries at least one known bad value (BITB58 listed at 22.22% against a corrected 0.77%). (8) **Multi-method deviation:** CoralNet points (12.5%) and Coralscapes masks (12.8%) sit ~12 percentage points below the formal-transect figure (24.6%; published deviation range −12.1 to −11.8 pp), i.e. the CV products recover about half the transect cover. Do not average the three figures; the transect value is ground truth for cover. See flag (3) for why the CV figures here are weaker than usual. (9) **The imagery and the transects disagree about non-coral invertebrates.** Large gorgonian sea fans are conspicuous on the wall section (P4300860, P4300862, P4300864) and a spiny lobster is photographed under a ledge (P4300852), yet published PITS is only 2.6% — the lowest of this batch. The most likely reconciliation is that the gorgonian-hung wall lies outside the transect footprint, but this is an inference, not a verified explanation, and it bears on how representative the transects are of the site as a whole. (10) ***Porites* is nearly absent here (0.87% of benthos, 4% of coral, 8th-ranked genus).** For comparison within the same atoll and the same survey week, BISA102 two days earlier had *Porites* at 14.11% holding 46% of its coral. The contrast is recorded as an observation; no causal explanation is offered and none is tested by these data. (11) **The `exposure-original` cell is BLANK** in `Chagos benthic data by site (n=109).csv`; the header reports `exposure-revised` = exposed. No class was invented. (12) **Three genera are effectively depth-restricted:** *Millepora* occurs only at D2 (2.00%, zero at D3 and D4), *Isopora* is essentially confined to D4 (3.10%, 0.20% at D3, zero at D2), and *Pachyseris* and *Psammocora* occur at D4 only. Site-level means for these genera describe no transect. (13) **Notable fauna:** three shark encounters in the sampled frames (P4300837, P4300841, P4300843) plus a fourth in the monochrome frame P4300839; a spiny lobster (P4300852). **No COTS, no disease, no *Drupella* and no predation scarring were observed** across the 20 frames examined — an explicit absence-of-evidence finding, bounded by the 19 frames not reviewed. (14) **Formal replication is `N Rows` = 3**, and there is no D1 transect, so the shallowest reef zone is unsampled. The depth codes D2–D4 carry no metre boundaries in `Chagos benthic data by site-depth (n=336 rows).csv`; they are ordered shallow-to-deep stations, and the 34.75 m in the header is the site-level survey depth, not the depth of any one transect. (15) **This dive shares its date (30 Apr 2015) with BIVB109 at Victory Bank.** The two folders do not overlap in frame numbers — BISA107 runs P4300837–876 and BIVB109 runs P4300936–993 — so no frames are ambiguous between the sites; the sequential numbering implies BISA107 was photographed earlier in the day, but the dive order was not independently confirmed. (16) **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 **7.08%, rank 58 of 110** (1st training: 7.08%, rank 57 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 BISA107 benthic survey data (CSV)](../benthic_data/BISA107_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).