Part of the regional summary: Indonesia - Maluku

Pulau Koon's second dive is a mound-and-channel slope of large coral bommies on open sand, headlined by a spectacular tabulate Acropora table over two meters across and a near-monospecific stand of rust-red digitate coral. A close-up series shows individual Acropora branches stark white against normally pigmented neighbors on the same colony; Anderson's dive notes record no bleaching. No crown-of-thorns starfish (COTS) or anchor damage seen; coral cover at this site is about 25% (blind Claude computer-vision point count).
Penyelaman kedua di Pulau Koon adalah lereng berpola gundukan dan kanal dengan bommie karang besar di atas pasir terbuka, dengan daya tarik utama meja Acropora tabulat yang spektakuler selebar lebih dari dua meter dan rumpun karang digitata merah karat yang hampir monospesifik. Serangkaian close-up menunjukkan cabang-cabang Acropora tertentu putih bersih di antara cabang berpigmen normal pada koloni yang sama; catatan selam Anderson tidak mencatat pemutihan. Tidak terlihat COTS atau kerusakan jangkar; tutupan karang di lokasi ini sekitar 25% (penghitungan titik buta oleh computer vision Claude).
Terjemahan ringkasan ini oleh AI; versi bahasa Inggris yang berlaku.
Eleven independent survey stations shown with the program’s logo — click one to see how far it is from this site.
Two other CRD dive sites with their own pages within 5 km shown in orange — click a marker to open its page.
6 of the 47 photographs on this page are shown beside the text that discusses them under "Site description" below.
Anderson's one surface frame (not shown in the reef gallery) establishes the setting: a low, flat, palm-fringed sand cay on a calm horizon under a broken sky, with the tender and dive gear in the foreground — a very different landform from the karst towers of Raja Ampat, and consistent with Pulau Koon's position out in the Banda Sea rather than in the Papuan islands. Underwater, this is a deep, current-exposed slope with a wide white carbonate sand plain at its foot. Visibility was reduced: Anderson's notes record about 8 m, murkier than the first dive, although strobe-lit frames still hold real color at close range.
The structure is mound-and-channel: discrete coral bommies, plate stacks, and rock outcrops rising off broad open sand, with a steeper coral-covered flank running up out of frame and a hazy blue drop-off falling away on the other side. Large areas of the substrate are bare sand or coralline-crusted dead pavement — this is not a wall-to-wall coral carpet, and several frames are mostly sand with a single colony in them. Hard corals cover the full range of growth forms and several colonies are genuinely large. Tabulate Acropora is spectacular: at least one table well over two meters across, its underside a dark cavern sheltering butterflyfish, with a clean fringe of pale branch tips around the rim (e.g., Image 37).
Alongside it are corymbose and digitate acroporids, delicate fine-branched staghorn thickets with lilac-blue branch tips, and — the visual signature of this dive — a huge, near-monospecific stand of a thick, blunt-branched, deep rust-red digitate coral with white branch tips, filling two wide frames end to end (e.g., Image 27).
Massive and encrusting corals include big dome and helmet Porites-form bommies with the fine even corallites clearly resolved in macro, plating and laminar colonies with cupped margins on open sand, and coarse meandroid/knobbly heads (e.g., Image 3).
Anderson's own identifications record Seriatopora caliendrum, a seriatoporid, and Stylophora and Seriatopora together — pocilloporids are clearly a focus of this dive and appear as fine bushy colonies in cream, white, and pale pink. Octocorals and other soft-bodied cover are abundant on the bommie flanks: dense white and pale-lilac carpets of fine feathery pulsing octocoral (Xenia/Anthelia mold) blanketing whole rock faces, cream and gray leathery soft corals, tall bushy soft-coral trees, and bright red-orange Dendronephthya-type clumps. Black-coral and hydroid bushes (feathery ochre and dark brown fans) are everywhere on the deeper slope, along with fine whip corals. Non-coral benthos is unusually prominent and worth recording: extensive purple, magenta, and pink crustose coralline algae, in places forming whole encrusted overhangs; brilliant red and scarlet encrusting sponges; a large maroon barrel/vase sponge on sand; small blue, yellow, white, and translucent solitary and colonial ascidians in dense clusters; a purple-and-yellow Polycarpa-pattern ascidian; a striking pink-and-cream flabellate sponge; abundant crinoids; and Christmas-tree worms embedded in massive Porites (e.g., Image 2).
Anderson shot a close-up series of at least five frames of branching Acropora in which individual branches and branch segments are stark white beside normally pigmented brown-gold branches on the same colony (e.g., Image 4).
Anderson's dive notes record no bleaching on this dive, and the frames alone cannot say whether the white segments are bleached tissue, recent tissue loss, or bare skeleton. Separately, the massive Porites frames show small bare notches and pale lesion patches, some occupied by a blenny and some colonized by pink coralline algae and worm tubes — a mix of normal borer/blenny holes and genuine small tissue-loss scars. No crown-of-thorns, no obvious disease band, and no anchor or blast damage were seen.
Fauna recorded directly in the frames: two moray eels — a large dark giant-moray-type under a coralline-encrusted overhang, and a boldly banded zebra/barred moray in a sand-floored cave; a crocodilefish lying flat and perfectly camouflaged on open sand; a large scorpionfish tucked under an Acropora table; a spotted garden eel standing out of the sand plain (Anderson photographed it twice); a black-and-orange-tipped phyllidiid nudibranch (Phyllidia-pattern) crossing coarse white sand; an emperor or large-eyed bream cruising the blue over the slope; a bicolor angelfish and yellow wrasses; dense clouds of red and orange anthias over every high point; damselfishes, chromis, butterflyfish, and a coral-dwelling blue-striped blenny that Anderson photographed twice, sitting in bare notches on a massive Porites (e.g., Image 5).
Ceccarelli et al. (2022) surveyed this site (Koon MPA) in 2016, 2018, and 2020. Within 5 km, eleven independent stations recorded station-year means of 18–47% hard coral cover between 2016 and 2023; the nearest, Institut Pertanian Bogor (IPB) at KOE02, 0.13 km away, read a mean of 26% in 2023.
Regional context, not a survey of this site: Razak et al. (2026) found no significant trend in hard coral cover at Koon, the marine protected area off eastern Seram (a rise of 0.2% per year at 17 sites, surveyed by point-intercept transect at about 8–10 m in 2016, 2018, and 2020 by Yayasan WWF Indonesia); this is a trend for that location, not this site’s cover. Koon warmed by 0.25°C per decade over 1985–2023, and the only NOAA bleaching alerts the study lists there ran from December 2022 to February 2023 (73 days at Bleaching Alert Level 1).
Pulau Koon, second dive site, 17 October 2018.
Man-made content: the tender and gear on the surface, one frame in which Anderson is holding a blue underwater slate with a written line against a coral colony (survey work in progress), and a gear bag reading "MSY SEAHO..." — the liveaboard's name — in the surface frame.
| Method | Hard coral cover (%) | Photos (n) ± SE |
|---|---|---|
| In-house DINOv3 | 19.4 | n = 53 · SE 1.6 |
| CoralSCOP edited | 32.6 | n = 53 · SE 3.3 |
| Claude (edited photos) | 24.7 | n = 52 · SE 4.1 |
The two computer-vision methods disagree by 13.2 pp (19.4–32.6%) and read on both sides of Claude's blind point count (24.7%); 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.
| Agency / program | Year(s) | Stations / distance | Link |
|---|---|---|---|
| Institut Pertanian Bogor (IPB) | 2023 | 6 stations, 0.13–4.44 km | Records ↓ |
| WWF-Indonesia / WWF-US | 2016–2020 | 5 stations, 0.13–4.44 km | Records ↓ |
Coral Reef DiagnosticsYear-average (illustrative only, programs differ)Within 5 km, 11 independent survey stations recorded hard coral cover in 21 surveys between 2016 and 2023. Of the 5 stations surveyed in more than one year, 3 declined (by 3.0–9.0 points), 1 increased (by 16 points), and 1 changed by less than 1 point between their first and latest surveys. Four stations share the longest span (2016–2020); one, KOE02 (WWF-Indonesia / WWF-US, 0.13 km away), went from 18% in 2016 to 34% in 2020. Since 2021, station-year means average 30% (range 19–38%; n = 6 station-years from 7 surveys). This site's own estimate (25%) sits 1 point below the recent range of station-year means up to 2 km away (26–38%), inside the 7-point method-noise floor (inconclusive); the methods differ, so the comparison is context, not a like-for-like test.
Collected by other programs using different methods, so these are context for the figures above, not a like-for-like comparison. Where a station was surveyed more than once, the cell shows either the range of its values or, as a single value, the mean of surveys made in one year. Station names are not dive-site names.
All images from Coral Reef Diagnostics. Data are from Coral Reef Diagnostics and the organizations named in the data sources above.
© Coral Reef Diagnostics.