Development of potential fishing ground estimation models by machine learning for understanding interspecies relationships

• Project/Area Number: 18K05803
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 40030:Aquatic bioproduction science-related
• Research Institution: Japan Agency for Marine-Earth Science and Technology
• Principal Investigator: Igarashi Hiromichi 国立研究開発法人海洋研究開発機構, 付加価値情報創生部門(情報エンジニアリングプログラム), 副主任研究員 (10578157)
• Co-Investigator(Kenkyū-buntansha): 阿部 泰人 北海道大学, 水産科学研究院, 助教 (40627246) ; 齊藤 誠一 北海道大学, 北極域研究センター, 研究員 (70250503)
• Project Period (FY): 2018-04-01 – 2021-03-31
• Project Status: Completed (Fiscal Year 2020)
• Budget Amount: ¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000); Fiscal Year 2020: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000); Fiscal Year 2019: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000); Fiscal Year 2018: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
• Keywords: カツオ / アカイカ / サンマ / ハビタットモデル / 魚種間関係 / 漁場推定 / 機械学習 / 黒潮親潮移行域

Outline of Final Research Achievements

We developed an integrated habitat model for estimating spatial distributions of three species, skipjack tuna (Katsuwonus pelamis), Pacific saury (Cololabis saira), and neon flying squid (Ommastrephes bartramii) in the western North Pacific, and examined their prey-predator relationships using this machine learning model. As the results, we found the statistical relationships with the distribution of these three species in the Kuroshio-Oyashio transition region and the integrated habitat model developed for estimating Skipjack tuna distribution using potential suitable area for Pacific saury shows good performances for a practical use of fishing ground exploration.

Academic Significance and Societal Importance of the Research Achievements

本研究では、個別魚種のハビタットモデルによる推定した分布を他の2魚種のモデルに取り込むことで複数魚種の相互相関から捕食・被食の重要性を定量的に明らかにし漁場推定・予測精度の向上を実現した点が学術的に意義がある。これにより、これまで活用されていなかった異なる魚種の分布情報を機械学習モデルに反映することで、漁業者から求められる、漁場探索で大きなコストとなる漁船の燃油消費を抑え安定的な収益確保を図る手段としての高精度な漁場推定を実現するモデル構築につながる要素技術が開発できた。

Research Products

• [Journal Article] Identification of skipjack tuna (Katsuwonus pelamis) pelagic hotspots applying a satellite remote sensing-driven analysis of ecological niche factors: A short-term run (2020)
  • Author(s): Mugo Robinson、Saitoh Sei-Ichi、Igarashi Hiromichi、Toyoda Takahiro、Masuda Shuhei、Awaji Toshiyuki、Ishikawa Yoichi
  • Journal Title: PLOS ONE Volume: 15 Issue: 8 Pages: 1-25
  • DOI: 10.1371/journal.pone.0237742
  • Peer Reviewed / Open Access / Int'l Joint Research
• [Journal Article] Ensemble Modelling of Skipjack Tuna (Katsuwonus pelamis) Habitats in the Western North Pacific Using Satellite Remotely Sensed Data; a Comparative Analysis Using Machine-Learning Models (2020)
  • Author(s): Mugo Robinson、Saitoh Sei-Ichi
  • Journal Title: Remote Sensing Volume: 12 Issue: 16 Pages: 2591-2591
  • DOI: 10.3390/rs12162591
  • Peer Reviewed / Open Access / Int'l Joint Research
• [Presentation] Development of integrated habitat suitability index models of skipjack tuna, Pacific saury and neon flying squid in the western North Pacific (2020)
  • Author(s): H. Igarashi, Y. Ishikawa, Y. Tanaka, S. Saitoh, H. Abe, Y. Imamura
  • Organizer: Ocean Science Meeting 2020
  • Int'l Joint Research
• [Presentation] A statistical approach to estimate optimal habitat suitability of skipjack tuna and neon flying squid in the western North Pacific (2018)
  • Author(s): Hiromichi Igarashi, Sei-ichi Saitoh, Fumihiro Takahashi, Yoichi Ishikawa, Masafumi Kamachi
  • Organizer: 14th Pan Ocean Remote Sensing Conference
  • Int'l Joint Research
• [Presentation] 東北沿岸～沖合域における海況モデルを活用した漁場推定～漁業者の利用をめざして～ (2018)
  • Author(s): 五十嵐弘道
  • Organizer: 第2回環境モニター報告会