Application of deep learning techniques for determining the spatial extent and classification of seagrass beds, Trang, Thailand

Takehisa Yamakita; Fumiaki Sodeyama; Napakhwan Whanpetch; Kentaro Watanabe; Masahiro Nakaoka

doi:10.1515/bot-2018-0017

Article

Application of deep learning techniques for determining the spatial extent and classification of seagrass beds, Trang, Thailand

Takehisa Yamakita
Takehisa Yamakita is a spatial ecologist at JAMSTEC. He was awarded a PhD by Chiba University for research on spatial dynamics of seagrass. As a postdoctoral researcher at the University of Tokyo, he began work to predict marine biodiversity over Japan and Asia to support conservation policy. Recently, he has worked as a sub-leader of the marine group in a project S-15; PANCES, as lead author of an IPBES Regional Assessment, as a unit sub-leader of tsunami assessment in Tohoku Ecosystem-Associated Marine Sciences (TEAMS). He also had lectures as an associate professor at Hiroshima University and a lecturer at Sophia University.
, Fumiaki Sodeyama
Fumiaki Sodeyama is a technical assistant staff at JAMSTEC and serves as an adjunct instructor/lecturer at Urawa University. He earned a master’s degree in developmental biology at the University of Tokyo. As a researcher at Misaki Marine Biological Station, he studied the life history and distribution of the feather star, including a genetic survey across Japan. He is currently working in the marine group of S-15 (Predicting and Assessing Natural Capital and Ecosystem Services) and testing several deep learning techniques.
, Napakhwan Whanpetch
Napakhwan Whanpetch is a marine ecologist at Kasetsart University. She was awarded a PhD by Chiba University for research on the variation of macrobenthic invertebrate communities in seagrass. She started her professional research career in the Department of Marine Science, Faculty of Fisheries, Kasetsart University, where she now serves as a lecturer and the Assistant Dean for Academic Affairs. She is part of the research program that monitors the impact on the Andaman Sea of the 2004 Indian Ocean earthquake and tsunami, as well as the program researching marine ecosystems at the Andaman Coastal Research Station for Development, Ranong Province, and adjacent areas.
, Kentaro Watanabe
Kentaro Watanabe is a researcher at the Waterfront Vitalization and Environment Research Foundation. He earned his master’s degree at the Marine Ecology Lab of Hokkaido University and studied the spatio-temporal variation of seagrass and seaweed using remote sensing, databases, and GIS at the Akkeshi Marine Station, Field Science Center for Northern Biosphere, Hokkaido University.
and Masahiro Nakaoka
Masahiro Nakaoka is a professor at Hokkaido University and the Director of the Akkeshi Marine Station, Field Science Center for Northern Biosphere, Hokkaido University. He completed his PhD and undergraduate studies at the University of Tokyo. His marine ecology research ranges from basic subjects like plant–animal interactions in seagrass beds to more applied topics like evaluating multiple ecosystem services of coastal ecosystems in Asia. He collaborates with researchers from other countries, using the Akkeshi Marine Station as a base for international research networks, including ZEN (Zostera Experimental Network), GAME (Global Approach by Modular Experiments) and TSUNAGARI (Belmont Forum–funded international project).

Published/Copyright: June 12, 2019

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information Explore this Subject

From the journal Botanica Marina Volume 62 Issue 4

Abstract

Few studies have investigated the long-term temporal dynamics of seagrass beds, especially in Southeast Asia. Remote sensing is one of the best methods for observing these dynamic patterns, and the advent of deep learning technology has led to recent advances in this method. This study examined the feasibility of applying image classification methods to supervised classification and deep learning methods for monitoring seagrass beds. The study site was a relatively natural seagrass bed in Hat Chao Mai National Park, Trang Province, Thailand, for which aerial photographs from the 1970s were available. Although we achieved low accuracy in differentiating among various densities of vegetation coverage, classification related to the presence of seagrass was possible with an accuracy of 80% or more using both classification methods. Automatic classification of benthic cover using deep learning provided similar or better accuracy than that of the other methods even when grayscale images were used. The results also demonstrate that it is possible to monitor the temporal dynamics of an entire seagrass area, as well as variations within sub-regions, located in close proximity to a river mouth.

Keywords: Andaman Sea; deep learning; land cover classification; long-term dynamics; remote sensing

About the authors

Takehisa Yamakita

Takehisa Yamakita is a spatial ecologist at JAMSTEC. He was awarded a PhD by Chiba University for research on spatial dynamics of seagrass. As a postdoctoral researcher at the University of Tokyo, he began work to predict marine biodiversity over Japan and Asia to support conservation policy. Recently, he has worked as a sub-leader of the marine group in a project S-15; PANCES, as lead author of an IPBES Regional Assessment, as a unit sub-leader of tsunami assessment in Tohoku Ecosystem-Associated Marine Sciences (TEAMS). He also had lectures as an associate professor at Hiroshima University and a lecturer at Sophia University.

Fumiaki Sodeyama

Fumiaki Sodeyama is a technical assistant staff at JAMSTEC and serves as an adjunct instructor/lecturer at Urawa University. He earned a master’s degree in developmental biology at the University of Tokyo. As a researcher at Misaki Marine Biological Station, he studied the life history and distribution of the feather star, including a genetic survey across Japan. He is currently working in the marine group of S-15 (Predicting and Assessing Natural Capital and Ecosystem Services) and testing several deep learning techniques.

Napakhwan Whanpetch

Napakhwan Whanpetch is a marine ecologist at Kasetsart University. She was awarded a PhD by Chiba University for research on the variation of macrobenthic invertebrate communities in seagrass. She started her professional research career in the Department of Marine Science, Faculty of Fisheries, Kasetsart University, where she now serves as a lecturer and the Assistant Dean for Academic Affairs. She is part of the research program that monitors the impact on the Andaman Sea of the 2004 Indian Ocean earthquake and tsunami, as well as the program researching marine ecosystems at the Andaman Coastal Research Station for Development, Ranong Province, and adjacent areas.

Kentaro Watanabe

Kentaro Watanabe is a researcher at the Waterfront Vitalization and Environment Research Foundation. He earned his master’s degree at the Marine Ecology Lab of Hokkaido University and studied the spatio-temporal variation of seagrass and seaweed using remote sensing, databases, and GIS at the Akkeshi Marine Station, Field Science Center for Northern Biosphere, Hokkaido University.

Masahiro Nakaoka

Masahiro Nakaoka is a professor at Hokkaido University and the Director of the Akkeshi Marine Station, Field Science Center for Northern Biosphere, Hokkaido University. He completed his PhD and undergraduate studies at the University of Tokyo. His marine ecology research ranges from basic subjects like plant–animal interactions in seagrass beds to more applied topics like evaluating multiple ecosystem services of coastal ecosystems in Asia. He collaborates with researchers from other countries, using the Akkeshi Marine Station as a base for international research networks, including ZEN (Zostera Experimental Network), GAME (Global Approach by Modular Experiments) and TSUNAGARI (Belmont Forum–funded international project).

Acknowledgments

We thank Chittima Aryuthaka, Yaowaluk Monthum, N. Tippamas Srisombat, Suwat Pleumarom, and other staff members of Kasetsart University and the Marine National Park Education Center, Thailand, for field assistance. We also thank Khanjanapaj Lewmanomont, Chatcharee Supanwanid, Sompoch Nimsantichareon, and Isao Koike for their invaluable support in various aspects of our field research, and Norihiro Yamate, Hiroo Imaki, and members of Pacific Spatial Solutions Inc. for their technical assistance. This study was partially supported by grants from the Japan Society for the Promotion of Science (Funder Id: http://dx.doi.org/10.13039/501100001691, nos. 07J02341, 11740425, 16405007) and the Environment Research and Technology Development Fund [S-15, Predicting and Assessing Natural Capital and Ecosystem Services (PANCES)] of the Ministry of the Environment, Japan. This study is dedicated to the memory of the late Dr. Chittima Aryuthaka, who was dedicated to organizing various international programs on marine ecology in Thailand. This research could not have been conducted without her support.

References

Asaad, I., C.J. Lundquist, M.V. Erdmann and M.J. Costello. 2017. Ecological criteria to identify areas for biodiversity conservation. Biol. Conserv. 213: 309–316.10.1016/j.biocon.2016.10.007Search in Google Scholar

Beckham, C. and C. Pal. 2017. A step towards procedural terrain generation with GANs. arXiv preprint arXiv:1707.03383.Search in Google Scholar

Campbell, J.B. and R.H. Wynne. 2011. Introduction to remote sensing. Guilford Press, NY, USA.Search in Google Scholar

Chansang, H. and S. Poovachiranon. 1994. Distribution and species composition of seagrass beds along the Andaman Sea Coast of Thailand. Phuket Marine Biological Center Research Bulletin (Thailand) 59: 43–52.Search in Google Scholar

Creswell, A., T. White, V. Dumoulin, K. Arulkumaran, B. Sengupta and A.A. Bharath. 2018. Generative adversarial networks: an overview. IEEE Signal Process. Mag. 35: 53–65.10.1109/MSP.2017.2765202Search in Google Scholar

Dahdouh-Guebas, F. 2002. The use of remote sensing and GIS in the sustainable management of tropical coastal ecosystems. Environ. Dev. Sustain. 4: 93–112.10.1023/A:1020887204285Search in Google Scholar

Department of Marine and Coastal Resources (DMCR). 2005. Rapid assessment of the tsunami impact on marine resources in the Andaman Sea, Thailand.Search in Google Scholar

Foga, S., P.L. Scaramuzza, S. Guo, Z. Zhu, R.D. Dilley, T. Beckmann, G.L. Schmidt, J.L. Dwyer, M. Joseph Hughes and B. Laue. 2017. Cloud detection algorithm comparison and validation for operational Landsat data products. Remote Sens. Environ. 194: 379–390.10.1016/j.rse.2017.03.026Search in Google Scholar

Fortes, M.D., J.L.S. Ooi, Y.M. Tan, A. Prathep, J.S. Bujang and S.M. Yaakub. 2018. Seagrass in Southeast Asia: a review of status and knowledge gaps, and a road map for conservation. Bot. Mar. 61: 269–288.10.1515/bot-2018-0008Search in Google Scholar

Goodfellow, I.J., Y. Bengio and A. Courville. 2016. Deep learning: adaptive computation and machine learning. MIT Press, Cambridge.Search in Google Scholar

Hemminga, M.A. and C.M. Duarte. 2000. Seagrass ecology. Cambridge University Press, Cambridge.10.1017/CBO9780511525551Search in Google Scholar

Faridah-Hanum, I., G.S. Rawat, T. Yahara, M. Abi-Said, R.T. Corlett, F. Courchamp, R. Dai, H. Freitag, T. Haryoko, C.L. Hewitt, T. Hussain, T. Kadoya, G. Maheswaran, T. Miyashita, B. Mohan Kumar, A. Mohapatra, T. Nakashizuka, J.J. Piggott, C. Raghunathan, R. Rawal, A. Sheppard, Y. Shirayama, Y. Son, N. Takamura, S. Thwin, T. Yamakita, C.M. Febria, A. Niamir. 2018. Chapter 3: Status, trends and future dynamics of biodiversity and ecosystems underpinning nature’s contributions to people. In: (M. Karki, S. Senaratna Sellamuttu, S. Okayasu, W. Suzuki, eds.) The IPBES regional assessment report on biodiversity and ecosystem services for Asia and the Pacific. Secretariat of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, Bonn, Germany. pp.175–264.Search in Google Scholar

Isola, P., J.-Y. Zhu, T. Zhou and A.A. Efros. 2016. Image-to-image translation with conditional adversarial networks. arXiv preprint arXiv:1611.07004.10.1109/CVPR.2017.632Search in Google Scholar

Kawaguchi, S. and K.-I. Hayashizaki. 2011. Biodiversity studies on seaweeds and sea grasses in the coastal waters of Southeast Asia (Project-3: Seaweed/seagrass Group). Coastal Marine Science in Southeast Asia – Synthesis Report of the Core University Program of the Japan Society for the Promotion of Science: Coastal Marine Science (2001–2010), pp. 49–57.Search in Google Scholar

Komatsu, T., C. Igararashi, K. Tatsukawa, M. Nakaoka, T. Hiraishi and A. Taira. 2002. Mapping of seagrass and seaweed beds using hydro-acoustic methods. Fisheries Sci. 68: 580–583.10.2331/fishsci.68.sup1_580Search in Google Scholar

Kramer, O. 2016. Machine learning for evolution strategies. Springer, Heidelberg.10.1007/978-3-319-33383-0Search in Google Scholar

Landis, J.R. and G.G. Koch. 1977. The measurement of obserber agreement for categorical data. Biometrics 33: 159–174.10.2307/2529310Search in Google Scholar

Lewmanomont, K., S. Deetae and V. Srimanobhas. 1996. Seagrasses of Thailand. Seagrass biology. Proceedings of an International Workshop: Rottnest Island, Western Australia. Faculty of Sciences, the University of Western Australia, Nedlands. pp. 21–26.Search in Google Scholar

Lewmanomont, K., C. Supanwanid and M. Nakaoka. 2000. Species composition of seagrasses at Haad Chao Mai National Park, Trang Province, Thailand. Kasetsart Univ. Fish. Res. Bull. 22: 21–22.Search in Google Scholar

Loveland, T.R., B.C. Reed, D.O. Ohlen, J.F. Brown, Z. Zhu, L. Yang and J.W. Merchant. 2000. Development of a global land cover characteristics database and IGBP DISCover from 1 km AVHRR data. Int. J. Remote Sens 21: 1303–1330.10.1080/014311600210191Search in Google Scholar

Lyons, M.B., S.R. Phinn and C.M. Roelfsema. 2012. Long-term land cover and seagrass mapping using Landsat and object-based image analysis from 1972 to 2010 in the coastal environment of South East Queensland, Australia. ISPRS J. Photogramm. Remote Sens. 71: 34–46.10.1016/j.isprsjprs.2012.05.002Search in Google Scholar

Lyzenga, D.R. 1978. Passive remote sensing techniques for mapping water depth and bottom features. Appl. Opt. 17: 379–383.10.1364/AO.17.000379Search in Google Scholar

Lyzenga, D.R. 1981. Remote sensing of bottom reflectance and water attenuation parameters in shallow water using aircraft and Landsat data. Int. J. Remote Sens. 2: 71–82.10.1080/01431168108948342Search in Google Scholar

Matsunaga, T., A. Hoyano and Y. Mizukami. 2000. Monitoring of coral reefs on Ishigaki Island in Japan using multitemporal remote sensing data. Proc. SPIE Hyperspectral Remote Sensing of the Ocean 4154: 212–222.10.1117/12.411677Search in Google Scholar

Nakaoka, M. and C. Supanwanid. 2000. Quantitative estimation of the distribution and biomass of seagrass at Haad Chao Mai National Park, Trang Province, Thailand. Kasetsart Univ. Fish. Res. Bull. 22: 10–22.Search in Google Scholar

Nakaoka, M., Y. Tanaka, H. Mukai, T. Suzuki and C. Aryuthaka. 2007. Tsunami impacts on biodiversity of seagrass communities in the Andaman Sea, Thailand. (1) Seagrass abundance and diversity. Publications of the Seto Marine Biological Laboratory. Special Publication Series 8: 49–56.10.5134/70911Search in Google Scholar

Olthof, I. 2017. Mapping seasonal inundation frequency (1985–2016) along the St-John River, New Brunswick, Canada using the landsat archive. Remote Sens. 9: 143.10.3390/rs9020143Search in Google Scholar

Orth, R.J., T.J.B. Carruthers, W.C. Dennison, C.M. Duarte, J.W. Fourqurean, K.L. Heck, A.R. Hughes, G.A. Kendrick, W.J. Kenworthy, S. Olyarnik, F.T. Short, M. Waycott and S.L. Williams. 2006. A global crisis for seagrass ecosystems. BioScience 56: 987.10.1641/0006-3568(2006)56[987:AGCFSE]2.0.CO;2Search in Google Scholar

Short, F.T., C.A. Short and A. Novak. 2018. Seagrasses. In: (C.M. Finlayson, R. Milton, C. Prentice and N.C. Davidson, eds) The Wetland book: II: distribution, description, and conservation. Springer, Dordrecht, Netherlands. pp. 73–91.Search in Google Scholar

Tittensor, D.P., C. Mora, W. Jetz, H.K. Lotze, D. Ricard, E.V. Berghe and B. Worm. 2010. Global patterns and predictors of marine biodiversity across taxa. Nature 466: 1098–1101.10.1038/nature09329Search in Google Scholar

United Nations. 2016. The first global integrated marine assessment. United Nations.10.1017/9781108186148Search in Google Scholar

Wabnitz, C. and S. Andréfouët. 2008. Regional-scale seagrass habitat mapping in the Wider Caribbean region using Landsat sensors: applications to conservation and ecology. Remote Sens. Environ. 112: 3455–3467.10.1016/j.rse.2008.01.020Search in Google Scholar

Waycott, M., C.M. Duarte, T.J.B. Carruthers, R.J. Orth, W.C. Dennison, S. Olyarnik, A. Calladine, J.W. Fourqurean, K.L. Heck, A.R. Hughes, G.A. Kendrick, W.J. Kenworthy, F.T. Short and S.L. Williams. 2009. Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proc. Natl. Acad. Sci. USA 106: 12377–12381.10.1073/pnas.0905620106Search in Google Scholar PubMed PubMed Central

Yaakub, S.M., J. Lean, S. Ooi, P. Buapet and R. Unsworth. 2018. Seagrass research in Southeast Asia. Bot. Mar. 61: 177–179.10.1515/bot-2018-0046Search in Google Scholar

Yamakita, T. 2018. Change of the ocean after the Great East Japan Earthquake -Utilization of Geographic Information System and GIScience. E-Res. Spec. Biol. 2: in press. in Japanease.Search in Google Scholar

Yamakita, T. 2019. Eelgrass beds and oyster farming at a lagoon before and after the Great East Japan Earthquake 2011: potential to apply deep learning at a coastal area. IGARSS 2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE. in press.10.1109/IGARSS.2019.8900354Search in Google Scholar

Yamakita, T. and M. Nakaoka. 2009. Scale dependency in seagrass dynamics: how does the neighboring effect vary with grain of observation? Popul. Ecol. 51: 33–40.10.1007/s10144-008-0119-zSearch in Google Scholar

Yamakita, T., K. Watanabe and M. Nakaoka. 2011. Asynchronous local dynamics contributes to stability of a seagrass bed in Tokyo Bay. Ecography 34: 519–528.10.1111/j.1600-0587.2010.06490.xSearch in Google Scholar

Yamakita, T., K. Sudo, Y. Jintsu-Uchifune, H. Yamamoto and Y. Shirayama. 2017. Identification of important marine areas using ecologically or biologically significant areas (EBSAs) criteria in the East to Southeast Asia region and comparison with existing registered areas for the purpose of conservation. Mar. Pol. 81: 273–284.10.1016/j.marpol.2017.03.040Search in Google Scholar

Yamakita, T., H. Yokooka, Y. Fujiwara, M. Kawato, S. Tsuchida, S. Ishibashi, T. Kurokawa and K. Fujikura. 2018. Image dataset of ophiuroid and other deep sea benthic organisms in 2015 extracted from the survey off Sanriku, Japan, by the research following the Great East Japan Earthquake 2011. Ecol. Res. 33: 285–285.10.1007/s11284-018-1571-7Search in Google Scholar

Article note

This article is related to special issue Seagrass research in Southeast Asia, published in Botanica Marina 2018, vol. 61, issue 3.

Received: 2018-02-23

Accepted: 2019-05-17

Published Online: 2019-06-12

Published in Print: 2019-08-27

You are currently not able to access this content.

Articles in the same Issue

https://doi.org/10.1515/bot-2018-0017

Keywords for this article

Andaman Sea; deep learning; land cover classification; long-term dynamics; remote sensing