BVI - Necker Island snorkel
More about this place: Necker Island (Virgin Limited Edition)
2026-07-02
More info
Virgin Islands (NOAA CRW virtual station)A shallow fringing reef snorkeled around Necker Island, where elkhorn coral still forms living stands at Red Dock and Turtle Beach but standing dead elkhorn and staghorn rubble are just as common. Fire coral, brain corals, and gorgonians fill in, and spiny lobsters and a shark resting on sand were seen.
Four sites at this location are mapped in teal — click a pin to open that site’s page.
Four other CRD dive sites with their own pages within 5 km shown in orange — click a marker to open its page.
These are unedited originals; edited versions will replace them.
28 of the 48 photographs on this page are shown beside the text that discusses them under "Site description" below.
Site description
Setting
These were shallow snorkels: the surface is in shot repeatedly, so most of the photographs were taken at roughly 1-5 m (e.g., Image 7).
General benthic features
The dominant impression is structure without much live tissue. Frame after frame shows standing dead acroporid framework: whole elkhorn (Acropora palmata) thickets and staghorn (A. cervicornis) rubble fields, gray and algae-filmed, over pavement coated in red turf, crustose coralline algae, and Dictyota-type brown macroalgae (e.g., Image 10, Image 17, Image 23, Image 24, Image 27).
Several wide shots hold no live stony coral at all (e.g., Image 22, Image 44).
Live cover is real but patchy and small-bodied. Live elkhorn (A. palmata) still forms multi-colony stands at Red Dock and Turtle Beach (e.g., Image 25), and a golden A. cervicornis thicket with white growing tips also appears.
Fire coral is the most conspicuous live element — bladed Millepora complanata with white margins, and encrusting M. alcicornis sheeting over dead framework (e.g., Image 29, Image 33, Image 38, Image 42).
Pale cream Porites porites finger clumps sprout from turf-covered massive mounds; mustard-yellow Porites astreoides lobes recur, repeatedly with pink coralline algae overgrowing dead margins (e.g., Image 13, Image 16, Image 26, Image 34).
Symmetrical brain corals (Pseudodiploria strigosa) appear in five frames, mostly intact — but one has half the colony dead and red-turfed, and one pale massive colony has a central patch of old dead skeleton crusted with pink crustose coralline algae (e.g., Image 2, Image 18, Image 30, Image 43).
Gorgonians carry the relief where coral does not: sea rods, sea plumes, and purple Gorgonia sea fans (e.g., Image 6, Image 31, Image 35, Image 39).
Coral cover
- Reference: Claude CV point count (this site's photos)
- CoralNet submission package
- Coral cover: 4.0% — a blind Claude computer-vision point count of 310 of Anderson's unedited photographs (30 points per photo), the unweighted mean of the four snorkel areas; each area's value is in the by-area table. Other computer-vision methods are listed in the method table.
Stressors
No tissue-loss lesions of the SCTLD kind were identifiable in any frame. One live elkhorn-type colony at Red Dock has a sharply bordered patch of recently exposed white skeleton at its center, between pigmented branches: recent partial mortality, cause not stated (e.g., Image 51).
Fish and other fauna
A nurse-shark-type shark resting on sand, two Caribbean spiny lobsters under ledges, a dark long-spined urchin, blue tangs, sergeant majors, bar jacks, brown chromis, and a maroon ball sponge (e.g., Image 3, Image 11, Image 17).
Comparisons with other data
We found no other surveys at this site as of 2026. If you would like to share yours, please email us.
Other notes
Anderson's five Necker snorkel folders hold 324 photographs. Some are topside or at the surface: the Red Dock pergola and pink jetty, the catamaran beach under a thick sargassum wrack line, Turtle Beach, and sand spits looking across to Virgin Gorda. Necker Island, BVI — circumnavigated over six dives/swims, 1-2 Jul 2026: Red Dock, Turtle Beach (×3 visits), Bali Hai-to-Turtle-Beach and Watersport-Beach-to-Bali-Hai transects. Positions of the individual areas are approximate.
% coral cover by site/region
Claude = blind Claude computer-vision point count (30 points per photo; the headline). CoralSCOP = mask-sum area index on the same areas, for comparison. All sites = unweighted mean of the area values, ± SD across areas.
| Area | Claude % coral cover | Claude std. dev. | Claude photos | CoralSCOP % coral cover | CoralSCOP std. dev. | CoralSCOP frames used | Frames available |
|---|---|---|---|---|---|---|---|
| Red Dock | 8.0% | ±12 | 87 | 20.8% | ±17 | 93 | 93 |
| Turtle Beach 2026-07-01, 2026-07-02 | 3.1% | ±5.6 | 83 | 16.9% | ±13 | 91 | 91 |
| Bali Hai (turtle) | 2.5% | ±6.3 | 89 | 12.1% | ±8.1 | 89 | 89 |
| Watersports / Bali Hai | 2.4% | ±4.9 | 51 | 12.3% | ±9.7 | 51 | 51 |
| All sites (n=4) | 4.0% | ±2.7 | 310 | 16% | ±4.2 | 324 | 324 |
CV model comparison
More info
Red Dock's higher value raises the mean; the median is the typical area. No computer-vision method yet estimates coral cover accurately on new images, so treat this as indicative rather than a measurement; it is revised as the methods improve.More info
DINOv3 and CoralSCOP were run on the same snorkel areas and are listed for comparison. No diver transect exists for this trip, so agreement between the rows is not accuracy, and no computer-vision method yet estimates coral cover accurately on new images; treat every row as indicative.| Method | Hard coral cover (%) | Photos (n) ± SE |
|---|---|---|
| In-house DINOv3 | 15.5 | n = 324 · SE 0.2 |
| CoralSCOP unedited | 15.5 | n = 324 · SE 0.2 |
| Claude | 4.0 | n = 310 · SE 0.2 |
The two computer-vision methods agree within 0.0 pp (15.5%) and read 11.5–11.5 pp above Claude's blind point count (4.0%); the headline is Claude's blind point count because no transect or manual count exists here.
See a more detailed, stand-alone version of this table.
Data & links
All images from Coral Reef Diagnostics. Data are from Coral Reef Diagnostics.
© Coral Reef Diagnostics.