Development of long-term monitoring method of agricultural land use dynamics in Japan using multiple satellite seonsor images

2018 Fiscal Year Final Research Report

• Project/Area Number: 16K07942
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Rural environmental engineering/Planning
• Research Institution: Kobe University
• Principal Investigator: Nagano Takanori 神戸大学, 農学研究科, 准教授 (70462207)
• Research Collaborator: Fujihara Yoichi 石川県立大学 ; Yoshikawa Natsuki 新潟大学
• Project Period (FY): 2016-04-01 – 2019-03-31
• Keywords: 合成開口レーダー / 農地 / 水土里ネット / 土地利用判別 / 水田 / マルチセンサー

Outline of Final Research Achievements

We have developed a technology that can determine the land use of agricultural land in Japan on a field basis in a wide range and inexpensively using of medium-resolution synthetic aperture radar satellite images and optical sensor satellite light images. By using field shape data provided by Midori-net together, rice grown and non-rice grown field distinction became possible with high accuracy by the end of June for fields larger than 0.25 ha with three synthetic aperture radar images of regular acquisition. While it was possible to distinguish soybean fields larger than 0.25 ha with additional use of visible light images, there were still problems in accurately determining abandoned fields.

Free Research Field

地域計画

Academic Significance and Societal Importance of the Research Achievements

現在日本において農地の動態調査は農業センサスが5年に1度行われている．一方農業土地利用は農家の高齢化や経済的な要因で近年大きく変化しつつあり，特に耕作放棄地が急速に増加している．農業土地利用の変化を把握し，適切な対策を講じるためには農地動態の調査が毎年行われることが望ましい．本研究の成果により全国レベルで農地土地利用が圃場単位で自動判別可能になる．現在補助金支給等の確認のため，市町村職員が目視で農地を悉皆調査しているが，その労働負荷削減にも大きく貢献する．