Utilization of ALOS data for mapping coastal habitats: examples of seagrass beds from boreal to tropical waters
Teruhisa Komatsu(1), Tatsuyuki Sagawa(1), Akihiko Mohri(1), Haifa Ben Romdhane(1), Masahiro Fukuda(1), Mahatma Lanuru(2), Mohd Ibrahim Seeni Mohd(3), Yoshihiko Sakanishi(4) and Thomas Belsher(5)
(1) The University of Tokyo, 1-15-1, Minamidai, Nakanoku, Tokyo 164-8639, Japan
(2) Hasanuddin University, Jalan Perintis Kamerdekaan Km 10, Makassar 90245, Indonesia
(3) Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia
(4) Hokkaido National Fisheries Research Institute, 116, Katsurakoi, Kushiro, Hokkaido 085-0802, Japan
(5) IFREMER, Station de Sète, Avenue Jean Monnet, BP 171, 34203 Sète, France
Sound coastal ecosystems provide important services such as habitats for animals and plants, and stabilizing effects of environments. They are indispensable for sustainable development of costal areas. However, human impacts such as fisheries or reclamation destroy coastal environments and ecosystems. Thus it is necessary to know present situation of coastal areas including coastal ecosystems and human activity to conserve or restore sound coastal ecosystems. It is desired to develop efficient mapping and monitoring systems of coastal areas. This study examines possibility of ALOS imagery for mapping different types of costal areas, habitats and fisheries facilities. For the first step of the study, we selected Akkeshi Lake in boreal waters as a study site because a broad seagrass bed in the blackish lake permits us to test the performance of AVNIR2 to detect seagrass distribution. Sea truth indicated that seagrass, Zostera marina, was distributed in shallow area in the lake. Image processing of AVNIR2 was done using supervised classification based on maximum likelihood method. The results show that AVNIR2 image will be applicable to map seagrass beds in shallow waters in boreal waters. For the second step, we selected Spermonde Islands in Sulawesi in Indonesia as a study site because typical tropical seagrass beds are distributed there. The image analysis conducted as the same method mentioned above successfully identified seagrass beds. These results indicate that ALOS AVNIR-2 imagery can be utilize to map seagrass beds in shallow waters from boreal to tropical waters.