Triton Bay - Little Komodo
Part of the regional summary: Triton Bay
More info
West Papua (NOAA CRW virtual station)Little Komodo's front side is dominated by a nearly uninterrupted forest of white soft coral down to 20 m, with staggering fish biomass including an enormous bumphead parrotfish and a wobbegong shark, plus scattered Acropora and abundant orange-and-green Tubastraea in the shallows. A blind Claude computer-vision point count puts coral cover around 12%.
Sisi depan Little Komodo didominasi hutan karang lunak putih yang nyaris tanpa putus hingga kedalaman 20 m, dengan biomassa ikan yang luar biasa termasuk ikan kakatua berbonggol raksasa dan hiu wobbegong, ditambah Acropora yang tersebar serta Tubastraea jingga-hijau yang melimpah di perairan dangkal. Penghitungan titik buta visi komputer Claude memperkirakan tutupan karang sekitar 12%.
Terjemahan ringkasan ini oleh AI; versi bahasa Inggris yang berlaku.
One other CRD dive site with its own page within 5 km shown in orange — click a marker to open its page.
5 of the 19 photographs on this page are shown beside the text that discusses them under "Site description" below.
Photo: Pierre7
Photo: Pierre8
Photo: Pierre9
Photo: Pierre10
Photo: Pierre11
Photo: Pierre12
Photo: Pierre13
Photo: Pierre14
Photo: Pierre15
Photo: Pierre16
Photo: Pierre17
Photo: Pierre18
Photo: Pierre19
Site description
Setting
This dive was on the front side of Little Komodo Island. It is very different from the island's other side. Visibility was 3-4 m.
General benthic features
At 15-20 m there was a forest of white soft coral in incredibly murky, low-light water; the descent landed on top of the soft coral garden, which held some soft corals never seen before (e.g., Image 2).
Soft coral forest was virtually uninterrupted with some barrel sponges and the odd small, hard coral colony (e.g., Image 1).
Orange and green Tubastraea was abundant at 5-7 m. Branchy plates and Acropora were relatively common at 5-7 m (e.g., Image 5).
There were some patches of urn tunicates in the shallows.
Coral cover
- Reference: Claude CV point count (this site's photos)
- Coral cover: 12% — a blind Claude computer-vision point count of 73 of Anderson's edited photographs (20 points per photo; water, fish, and obstructions excluded). Other computer-vision methods are listed in the method table.
Fish and other fauna
Fish biomass was much higher than on the island's other side, with an enormous bumphead parrotfish (Anderson’s notes: “one of the biggest I’ve seen”). Little yellow sea cucumbers were everywhere. An enormous angelfish was seen, and the soft coral garden ended at 20 m, with just rubble below. Several large groupers were seen. A wobbegong (carpet shark), Anderson's first, was a highlight of the dive (e.g., Image 3).
Fish biomass at 5-7 m was staggering (e.g., Image 4).
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.
Regional context, not a survey of this site: Bawole et al. (2026) report mean hard coral cover at 24 University of Papua monitoring sites in the Kaimana marine protected area around Triton Bay, surveyed by point-intercept transect at 8–10 m: 17%, 24%, and 32% in the no-take zones (eight sites) and 14%, 18%, and 13% in the use zones (16 sites) in 2012–2013, 2015–2016, and 2019. Cover differed significantly between zones but not over time (p = 0.828); these are zone means, not this site’s cover.
Dive record
This was the second dive of the day; Anderson's notes call it a great dive.
CV model comparison
More info
A few colony-focused photos raise the mean; the median is the typical photo. 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 site's photographs (see each row's n) 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 | 22.9 | n = 75 · SE 1.2 |
| CoralSCOP edited | 31.2 | n = 75 · SE 2.1 |
| Claude (edited photos) | 11.7 | n = 73 · SE 2.2 |
The two computer-vision methods differ by 8.3 pp (22.9–31.2%) and read 11.2–19.5 pp above Claude's blind point count (11.7%); 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.
Data & links
Images 6–18: Pierre; all other images from Coral Reef Diagnostics. Data are from Coral Reef Diagnostics.
© Coral Reef Diagnostics.
References
- Bawole R, Sala R, Wurarah RN, Tebaiy S, Sirfefa I, Kolibongso D, et al. (2026). Integrated assessment of coral reefs and reef fish assemblages in Kaimana Marine Protected Areas, West Papua. Jurnal Teknologi Perikanan dan Kelautan 17(2):243-252. DOI