Atmospheric and Hydrospheric Sciences (A)
Session Sub-categoryComplex & General(CG)
Session IDA-CG36
TitleEarth & Environmental Sciences and Artificial Intelligence
Short TitleEarth & Environmental Sciences and AI
Main Convener NameTomohiko Tomita
AffiliationFaculty of Advanced Science and Technology, Kumamoto University
Co-Convener 1NameKen-ichi Fukui
AffiliationOsaka University
Co-Convener 2NameDaisuke Matsuoka
AffiliationJapan Agency for Marine-Earth Science and Technology
Co-Convener 3NameSatoshi Ono
AffiliationKagoshima University
Session LanguageJ
ScopeIn recent years, we have been required to explore the gigantic environmental data leading global environmental studies of modern meteorology, oceanography, and hydrology to accomplish Society 5.0 and Sustainable Development Goals (SDGs). Such environmental data, which are typical "big data," include the reliable long-term observatory data, ground radar data, satellite observation, oceanographic observation, global objective analysis data, and so on. However, it is hard to say that such big data are fully utilized, or nobody may have examined many of them. Therefore, to examine the global environment more faithfully, we need to apply the techniques of artificial intelligence/machine learning on such big data, that is, spatiotemporal data modeling of artificial intelligence, prediction and detection by machine learning, techniques of automated data mining, and so forth. Only cross-cutting diagnosis of the gigantic environmental data could resolve a wide variety of environmental problems including measures against global warming. In addition, the application of artificial intelligence/machine learning on the big data would raise us to the 4th paradigm of science (the 1st paradigm, observation/experiment; 2nd, theory; 3rd, simulation; 4th, data driven). This session sincerely invites various research initiatives that accelerate the application of artificial intelligence/machine learning on such gigantic environmental data.
Presentation FormatOral and Poster presentation
Invited AuthorsShinichi Shirakawa (Yokohama National University)
TimePresentation NoTitlePresenterAbstract
Oral Presentation May 30 AM2
10:50 - 11:20ACG36-01Recent Progress in Deep Neural Network Architectures and their ApplicationsShinichi Shirakawa
11:20 - 11:35ACG36-02Improving performance of oxidant prediction using machine learning by optimizing input data based on atmospheric science knowledgeTomohiro Sato
11:35 - 11:50ACG36-03Spatiotemporal Climate Forecasting with ConvLSTMEkasit Phermphoonphiphat
11:50 - 12:05ACG36-04Deep-Learning Based Short-Term Prediction Method for Sea Ice Concentration Using Explanatory VariablesIssei Kawashima
Oral Presentation May 30 PM1
13:45 - 14:00ACG36-05Remote sensing of cloud using the deep learning based on three-dimensional radiative transferHironobu Iwabuchi
14:00 - 14:15ACG36-06Time series of cloud cover during the night derived from an omnidirectional cameraKarin miyabe
14:15 - 14:30ACG36-07Time series prediction of cloud cover using whole-sky images and meteorological elementsKazunori Ogohara
14:30 - 14:45ACG36-08Fundamental investigation about the inductive prediction of precipitation based on the time series dataNaoki Matsumoto
14:45 - 15:00ACG36-09Identification of ships image on SAR data using machine learningTakashi Sonoke
Presentation NoTitlePresenterAbstract
Poster Presentation May 30 PM2
ACG36-P01Automatic Detection of Illegal Waste Sites in UAS OrthoimagesYu-Ching Lin
ACG36-P02Using machine learning models to predict the benefits of solar power generation in northern TaiwanSHIH-CHIAO TSAI
ACG36-P03Change Point Detection and Visualization of Region of Interests on Weather Time Series Data Using Three-dimensional Convolutional Neural NetworkSatoshi Ono
ACG36-P04Estimation of sediment disaster risk around the Fuji River Basin using a machine learning method that considers trigger and inherent factorsKazuyoshi Souma
ACG36-P05Application of deep learning techniques to precipitation guidance in short-term weather predictionsTAKERU KURAKAMI
ACG36-P06Characteristics of precipitation in Japan evaluated from the ratio of precipitation eventsKenta Ogino