Algorithm development for automated deforestation mapping via integrated use of multiple satellite data

2020 Fiscal Year Final Research Report

• Project/Area Number: 19K24395
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 1101:Environmental analyses and evaluation, environmental conservation measure and related fields
• Research Institution: National Institute of Advanced Industrial Science and Technology
• Principal Investigator: Mizuochi Hiroki 国立研究開発法人産業技術総合研究所, 地質調査総合センター, 研究員 (20849963)
• Project Period (FY): 2019-08-30 – 2021-03-31
• Keywords: 森林消失検知 / 衛星データ / 機械学習

Outline of Final Research Achievements

This study aims to deforestation mapping via integrated use of multiple satellite datasets that have physically different features. In addition to the common satellite images detecting solar reflection from the earth surface (i.e., optical images), microwave images were utilized, which enable us to observe the worldwide forest even under the cloudy condition. Tropical forest sites (Cambodia and Peru) were selected and monitored by satellite images for past twenty-years, and deforestation maps were created for each image. Those maps were integrated via a machine-learning approach, resulting in temporally consistent deforestation maps.

Free Research Field

環境学

Academic Significance and Societal Importance of the Research Achievements

世界全体の森林面積は減少を続けており、とくに熱帯地域で大規模な森林減少が報告されている。開発された手法で、雲に覆われがちな熱帯林においても、従来よりも高頻度な定期観測が実現しうる点で、気候変動・生物多様性・水循環など様々な関連分野における社会的意義がある。また、複数衛星データの柔軟な統合手法は、今後ますます各国の衛星データがアーカイブ化・オープンフリー化されていく流れのなかで、衛星観測技術の発展にも資する学術的意義を有する。