The coastline of Cambodia stretches along the Gulf of Thailand including 69 islands. It supports rich diversity of marine
species. Distributions of habitats including mangroves, coral reefs and seagrass beds along the pristine Cambodian coast
still remains unknown compared to neighboring countries, Thailand and Vietnam. Cambodian seagrass beds form
habitats with rich biodiversity and economical value through marine ecosystem services playing as a key role against
climate change by reserving large amount of carbon. However, the general status of these seagrass habitats is poorly
researched and documented. Satellite image of Advance Visible and Near Infrared Radiometer type 2 (ALOS AVNIR-2)
with high resolution (10×10m) provides good information for seagrass habitat mapping. Study site was selected around
Rabbit (Koh Tonsay) Island with area of 2 km<sup>2</sup>. The objectives of this study are (1) to know spatial distribution of
seagrass beds around this island by ground survey and (2) map seagrass beds using the ALOS AVNIR-2 image with
ground truthing data. Ground truth survey was conducted in June 2011. Surveys along three transect lines revealed 8
species of seagrasses belonging to<i> Hydrocharitaceae </i>(4 species) and <i>Cymodoceaceae</i> (4 species) around the island. We
analyzed ALOS AVNIR-2 taken on 22 December 2009 to map distribution of seagrass beds around Koh Tonsay Island.
Results showed that remote sensing using ALOS AVNIR-2 data provides a practical tool for mapping seagrasses beds
around the island and information for future management and conservation of seagrass beds in Cambodia.
Sargassum species grow on rocks and dead corals and form dense seaweed beds. Sargassum beds play ecological roles
such as CO2 uptake and O2 production through photosynthesis, spawning and nursery grounds of fish, feeding ground for
sea urchins and abalones, and substrates for attached animals and plants on leaves and holdfasts. However, increasing
human impacts and climate change decrease or degrade Sargassum beds in ASEAN countries. It is necessary to grasp
present spatial distributions of this habitat. Thailand, especially its coastal zone along the Gulf of Thailand, is facing
degradation of Sargassum beds due to increase in industries and population. JAXA launched non-commercial satellite,
ALOS, providing multiband images with ultra-high spatial resolution optical sensors (10 m), AVNIR2. Unfortunately,
ALOS has terminated its mission in April 2011. However, JAXA has archived ALOS AVNIR2 images over the world.
They are still useful for mapping coastal ecosystems. We examined capability of remote sensing with ALOS AVNIR2 to
map Sargassum beds in waters off Sattahip protected area as a natural park in Chon Buri Province, Thailand, threatened
by degradation of water quality due to above-mentioned impacts. Ground truth data were obtained in February 2012 by
using continual pictures taken by manta tow. Supervised classification could detect Sargassum beds off Sattahip at about
70% user accuracy. It is estimated that error is caused by mixel effect of bottom substrates in a pixel with 10 x 10 m. Our
results indicate that ALOS AVNIR2 images are useful for mapping Sargassum beds in Southeast Asia.
Coastal habitats having high productivity provide numerous ecological services such as foods, protection from strong
waves through buffering effect, fixation of CO2 through photosynthesis, fostering biodiversity etc. However, increasing
human impacts and climate change decrease or degrade coastal habitats. ASEAN region is developing most rapidly in
the world. In the developing region, it is necessary to grasp present spatial distributions of habitats as a baseline data with
standardized mapping methods. Remote sensing is one of the most effective methods for mapping. Japan Aerospace
Exploration Agency (JAXA) provides non-commercial satellite images with ultra-high spatial resolution optical sensors
(10 m), AVNIR2, similar to LANDSAT TM. Using ALOS AVNIR2 images it may be possible to make habitat map in
the region. In Thailand, shrimp ponds cause degradation of coastal ecosystem through cutting mangroves and
eutrophicated discharge from ponds. We examined capability of remote sesing with ALOS AVNIR2 to map seagrass
beds in Khung Kraben Bay, Chanthaburi Province, Thailand, surrounded by shrimp ponds. We analyzed ALOS AVNIR2
taken on 25 January 2008. Ground truth survey was conducted in October 2010 using side scan sonar and scuba diving.
The survey revealed that there were broad seagrass beds consisting of Enhalus acroides. We used a decision tree to
detect seagrass beds in the bay with quite turbid seawater coupled with Depth-Invariant Index proposed by Lyzenga
(1985) and bottom reflectances. We could succeed to detect seagrass beds. Thus it is concluded that ALOS AVNIR2 is
practical to map seagrass beds in this region.