Indonesia - Ambon: Air Manis (Laha)

Setting

Air Manis is muck-with-patch-reef topography: a sand/silt/rubble bottom interspersed with raised coral bommies and low ridges carrying surprisingly dense live cover — near-100% Acropora/Porites carpets on the bommie tops against bare sand between them (e.g., Image 1).

1

Visibility ran moderate-to-good over the bommies, hazier and greener over the flatter muck.

General benthic features

Taxa in Anderson's photo names include table and staghorn Acropora, Montipora (with a question mark), Euphyllia, Seriatopora caliendrum, a fluorescent-green Stylophora, Pocillopora meandrina, and large Porites domes; the frames also show brain corals, fire coral, and soft leather corals (e.g., Image 2).

2

Coral cover

• Reference: CoralSCOP computer vision (this site's photos)
• Coral cover — completed photographic analysis.
• CoralSCOP processed all 72 photographs and returned a mean computer vision estimate of 31% (SD 23; median=26%).
• That number is not literal site-wide benthic cover: these are selected dive photographs mixing wide reef scenes, coral portraits, and fishing-impact documentation, and individual outputs range from 0.8% on a visibly coral-covered scene to 127% on a close-up.
• The impossible value above 100% confirms overlapping masks, while the very low values on obvious coral carpets show under-segmentation.
• The 31% result is therefore shown as a provisional photographic index, consistent with the visibly patchy habitat, and not as transect-equivalent coral cover, until an independent point-count or second image method is available.
• Coral cover: 31% — a single-method photographic index. This was a muck dive: the photographs pool coral patches with sand, silt, and rubble bottom that is not reef, so the value describes that mix, not the cover of a reef.
• The figure is a CoralSCOP computer vision over Anderson's own photographs (72 frames), with no second method to cross-check it.
• CoralSCOP sums retained mask fractions with overlaps possible and segments coral-shaped objects rather than living tissue, so it scores standing dead skeleton as coral and does not measure cover the way a transect does; against in situ transects it reads high.
• One frame in this set scored 127%, which is impossible as a cover value and is a mask-overlap artifact of the method — a caution about the index, not about the reef.

Stressors

Human-impact flag. At least five frames show a fine monofilament gillnet draped across coral bommies and the water column, with small fish visibly trapped in it — clear evidence of active or recently set fishing gear smothering part of the reef. One soft (leather) coral colony has a single lobe sharply bleached bright white against otherwise healthy tan-olive tissue; the bleaching is limited to that one colony. No crown-of-thorns starfish was seen. One isolated foliose colony on open sand shows an advancing tissue-loss lesion (a photo observation, not a diagnosis). Its white band of bare skeleton runs between live tissue and dark, colonized skeleton. Plating colonies in one frame carry white patches, cause unclear (e.g., Image 30).

30

A cushion star (Culcita, Anderson's identification), a corallivore, rests on rubble among small colonies, and a mushroom coral beside it is white on top, possibly a feeding scar (e.g., Image 51).

51

Fish and other fauna

Dense schools of chromis and fusiliers swarm the bommies alongside anthias, butterflyfish, and abundant crinoids (e.g., Image 3, Image 4, Image 5).

345

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.

Dive record

The first of three muck dives near Laha on Ambon's south coast, 2018-04-22.