Multidisciplinary and Interdisciplinary(M)
Session Sub-categoryApplied Geosciences
Session IDM-AG32
Related FieldsS, P, A
TitleMarine Earth Informatics
Short TitleMatine Earth Informatics
Main ConvenerNameSeiji Tsuboi
AffiliationJAMSTEC, Center for Earth Information Science and Technology
Co-Convener 1NameKeiko Takahashi
AffiliationJapan Agency for Marine and Earth Science and Technology
Co-Convener 2NameMasaki Kanao
AffiliationNational Institute of Polar Research
Session LanguageEJ
ScopeIn advancing the research of marine Earth science, observation and computer simulation is an essential element. In recent years, the performance of the observation apparatus is dramatically improved, along with the means of observation is diversified. It is becoming possible to observe in a resolution, which was not imaginable so far. Such data to be generated from the observation is tremendously large in quantity and its quality is drastically improved. To handle these huge and high quality dataset for data analysis, we need to have a high speed and large memory computer system but such a system now becomes within reach in our hands by the recent dramatic improvement of high performance computer system. On the other hand, researchers who can use this kind of large-scale computer in their studies are still quite limited. In this session, we try to review the situation of observation data that has undergone a dramatic change regarded with both quality and quantity in recent years of marine Earth science research. We also try to review the situation from a professional standpoint of simulation about the status of the high performance computer system to analyze these 'big data'. Also we focus on the state of the art data analysis technique and aim to share the outlook from the professional standpoint of computational science and professional position of observation science about the future direction of the marine Earth informatics research.
Presentation FormatOral and Poster presentation
Invited Authors
  • Tomoyuki Higuchi (The Institute of Statistical Mathematics)
  • Kazuhiro Yamasaki (NVIDIA)
Time Presentation No Title Presenter Abstract
Oral Presentation May 23 PM1
13:45 - 14:05 MAG32-01Visualization technology for machines to understand Bigdata: Automatic selection of feature vectors and barbarization of data analysis methodsTomoyuki Higuchi Abstract
14:05 - 14:20 MAG32-02Utilization of Deep Learning in mapping of the ocean floor: Extraction of brittle stars by image recognition, seagrass distribution using image to image translationTakehisa Yamakita Abstract
14:20 - 14:35 MAG32-03Development of Drilling Data Acquisition System and Attempt of Anomaly Detection from Drilling Data with Machine LearningTomoya Inoue Abstract
14:35 - 14:50 MAG32-04Scientific Visualization of Climate Simulation Data for Deep Convolutional Neural NetworkDaisuke Matsuoka Abstract
14:50 - 15:10 MAG32-05Recent advances of deep learning for HPC and GPU computer systemsKazuhiro Yamasaki Abstract
Oral Presentation May 23 PM2
15:30 - 15:45 MAG32-06To extract “right” information from a huge marine biodiversity information poolHosono Takashi Abstract
15:45 - 16:00 MAG32-07Multiscale-Multilayer Data Assimilation System for Smart Weather ForecastingOnishi Ryo Abstract
16:00 - 16:15 MAG32-08A study on the thermal environment of "Yato" area in hilly city -Analyzing air temperature and wind distribution by observation and numerical calculation-Makoto Yokoyama Abstract
16:15 - 16:30 MAG32-09Thermal environment in sea faced districtTooru Sugiyama Abstract
16:30 - 16:45 MAG32-10Ocean Data Publication for Broad Utilization in JAMSTECTomoki Sasaki Abstract
16:45 - 17:00 MAG32-11Seismic observations at Syowa Station and surrounding region of Antarctica - Sciece targets and data management for long-term monitoring -Masaki Kanao Abstract
Presentation No Title Presenter Abstract
Poster Presentation May 23 Core Time
MAG32-P01 Near real-time forecasts using global nonhydrostatic model on the Earth Simulator during intensive observations. Mikiko Ikeda Abstract
MAG32-P02 Detection and identification of multiple-type earthquakes based on deep learning approach Masaru Nakano Abstract
MAG32-P03 Earth Sciences big data analysis using Unsupervised Deep Learning, and challenge to "Earth Search". Daisuke Sugiyama Abstract
MAG32-P04 Data visualization service for the earth science Koji Imai Abstract
MAG32-P05 CAVELibWrapper: Development of a CAVELib Compatible Library for HMD-type VR Systems Shintaro Kawahara Abstract
MAG32-P06 Seamless Visualization of Weather Forecast Information with Nested Structure on Digital Globe Shintaro Kawahara Abstract
MAG32-P07 Geological Evolution of JAMSTEC DARWIN Database Takayuki Tomiyama Abstract
MAG32-P08 Quasi-Real-Time Surface Current Information of the Eastern Tsugaru Strait via Ocean Radar data Site "MORSETS" Hiroki Horikawa Abstract