Part of the regional summaries: Indonesia - Halmahera and Indonesia - Maluku

Jerot was dived in unusually cold water that had Anderson wondering in the log whether this is an upwelling site, and the reef runs a clean depth gradient — sparse knolls on a deep sand plain giving way to a high-cover shallow crest of Acropora, Pocillopora, and leather corals. A blind Claude computer-vision point count puts coral cover around 23%.
Jerot diselami dalam air yang luar biasa dingin, yang membuat Anderson bertanya-tanya dalam catatannya apakah ini lokasi upwelling, dan terumbunya memperlihatkan gradien kedalaman yang jelas — gundukan-gundukan renggang di dataran pasir dalam yang berganti menjadi puncak terumbu dangkal bertutupan tinggi berisi Acropora, Pocillopora, dan karang kulit. Penghitungan titik buta visi komputer Claude memperkirakan tutupan karang sekitar 23%.
Terjemahan ringkasan ini oleh AI; versi bahasa Inggris yang berlaku.
These are unedited originals; edited versions will replace them.
This page shows 50 of the 175 photographs from this site, including every one the text cites.
40 of the 50 photographs on this page are shown beside the text that discusses them under "Site description" below.
Jerot is a single continuous slope worked from a deep sand floor up onto a sunlit reef crest, and the imagery is organized almost perfectly by depth. Anderson's field note for the dive reads “super cold, weird thermoclines. Is this actually an upwelling area?” — the log records 22–25°C on a 22.1 m dive.
The deep base is a pale sand-and-rubble plain over crustose-coralline pavement, carrying isolated knolls with wide bare gaps between them; these supply most of the low-cover frames (e.g., Image 6, Image 13, Image 15, Image 17, Image 23, Image 33, Image 34, Image 41).
Growth there is a mix of broad laminar plates (Turbinaria- and Echinophyllia-type), green and cream tabulate Acropora sitting free on sand, foliose Montipora thickets with pale margins, maroon and violet encrusting sponges, barrel sponges (Xestospongia-type), and orange and lilac gorgonian fans (e.g., Image 1, Image 3, Image 8, Image 10, Image 14, Image 28, Image 31).
The mid-slope lifts into a very different reef: large intact Acropora tables up to a couple of meters across, Pocillopora and digitate Acropora heads, big domed massives, including a large dome with a white meandering wall pattern, and smooth lobate Porites, and dense crimson and salmon Dendronephthya soft coral (e.g., Image 9, Image 27, Image 29, Image 30, Image 37).
The crest and reef flat (the last stretch of the dive) is the high-cover half of the site: near-continuous Acropora, Pocillopora, and Seriatopora thickets over pavement, mixed with Sarcophyton and Lobophytum leather corals, feathery Xenia-type pulsing soft coral, and encrusting Montipora, with anthias, damsels, and anemonefish over almost every frame (e.g., Image 4, Image 5, Image 16, Image 20, Image 21, Image 24, Image 25, Image 32, Image 35, Image 38, Image 42, Image 45, Image 48, Image 49, Image 50, Image 51, Image 52, Image 54, Image 55).
Cover is highest in this shallow band.
Several branching Acropora colonies on the crest read conspicuously cream-white beside normally pigmented olive, lavender, and tan neighbors, with tissue intact and no turf algae or exposed skeleton. Whether these are paling colonies or naturally pale cream morphs cannot be resolved from the photographs, and the cold 22–25°C water logged on this dive argues against thermal bleaching, so they are recorded as unresolved, not as a bleaching observation. One rubble patch of broken staghorn-type Acropora fragments lies loose and overgrown by hydroids and coralline algae, consistent with old physical breakage of unknown cause (e.g., Image 2).
We found no other surveys at this site as of 2026. If you would like to share yours, please email us.
| Method | Hard coral cover (%) | Photos (n) ± SE |
|---|---|---|
| In-house DINOv3 | 27.5 | n = 172 · SE 0.9 |
| CoralSCOP unedited | 41.6 | n = 172 · SE 1.3 |
| Claude | 22.9 | n = 172 · SE 1.4 |
The two computer-vision methods disagree by 14.1 pp (27.5–41.6%) and read 4.6–18.7 pp above Claude's blind point count (22.9%); 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.
All images from Coral Reef Diagnostics. Data are from Coral Reef Diagnostics.
© Coral Reef Diagnostics.