Monitoring Mangrove Forest Changes in Cat Ba Biosphere Reserve Using ALOS PALSAR Imagery and a GIS-Based Support Vector Machine Algorithm Tien Dat Pham 1,2(&) , Kunihiko Yoshino 3 , and Naoko Kaida 4 1 Graduate School of Systems and Information Engineering, The University of Tsukuba, 1-1-1 Tennodai, Tsukuba 305-8573, Ibaraki Prefecture, Japan dat6784@gmail.com 2 Center for Agricultural Research and Ecological Studies (CARES), Vietnam National University of Agriculture (VNUA), Trau Quy, Gia Lam, Hanoi, Vietnam tiendat@cares.org.vn 3 Department of Biological and Environmental Engineering, Faculty of Agriculture, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-Ku, Tokyo 113-8657, Japan asky@mail.ecc.u-tokyo.ac.jp 4 Faculty of Engineering, Information, and Systems, The University of Tsukuba, 1-1-1 Tennodai, Tsukuba 305-8573, Ibaraki Prefecture, Japan naoko.kaida@sk.tsukuba.ac.jp Abstract. Cat Ba is one of the most well-known islands located in North Vietnam, which has been recognized as a biosphere reserve by United Nations Educational, Scientific and Cultural Organization (UNESCO) since 2004. Despite the large potential carbon stocks in mangrove forests of Cat Ba, the mangrove ecosystem of this island has suffered severe deforestation and forest degradation due to the conversion to shrimp aquaculture. Monitoring mangrove forest changes plays an important role for effective mangrove conservation and management. The objectives of this study were to map the spatial distribution of mangrove forest and to assess their changes between 2010 and 2015 in Cat Ba Biosphere Reserve, Hai Phong city of Vietnam using ALOS PALSAR data and a GIS-based support vector machine algorithm. For this purpose, ALOS PALSAR imagery for the above period and GIS data were collected. Then, spatial distributions of mangroves were derived using the support vector machine classifier. The results showed that the ALOS-2 PALSAR for 2015 achieves the overall accuracy of 85% and the kappa coef ficient of 0.81, com- pared with those of 81% and 0.77, respectively from the ALOS PALSAR for 2010. The mangrove forest areas in the Cat Ba Biosphere Reserve, Vietnam decreased by 15% from 2010 to 2015. This research shows the potential use of ALOS PALSAR data combined with machine learning techniques in monitoring mangrove forest changes in tropical and semi-tropical climates. Keywords: ALOS PALSAR  Cat Ba biosphere reserve  Hai Phong city  Mangrove changes  Support vector machines © Springer International Publishing AG 2018 D. Tien Bui et al. (eds.), Advances and Applications in Geospatial Technology and Earth Resources, https://doi.org/10.1007/978-3-319-68240-2_7