@InProceedings{MarcheziDSJDSDAWLZ:2022:NeNeAp,
               author = "Marchezi, Jos{\'e} Paulo and Dai, Lei and Silva, Ligia Alves da 
                         and Jauer, Paulo Ricardo and Dal Lago, Alisson and Sibcek, David 
                         and Deggeroni, Vin{\'{\i}}cius and Alves, Livia Ribeiro and 
                         Wang, Chi and Li, Hui and Zhengkuan, Liu",
          affiliation = "{Chinese Academy of Sciences (CAS)} and {Chinese Academy of 
                         Sciences (CAS)} and {Instituto Nacional de Pesquisas Espaciais 
                         (INPE)} and {Instituto Nacional de Pesquisas Espaciais (INPE)} and 
                         {Instituto Nacional de Pesquisas Espaciais (INPE)} and {NASA GSFC} 
                         and {Instituto Nacional de Pesquisas Espaciais (INPE)} and 
                         {Instituto Nacional de Pesquisas Espaciais (INPE)} and {Chinese 
                         Academy of Sciences (CAS)} and {Chinese Academy of Sciences (CAS)} 
                         and {Chinese Academy of Sciences (CAS)}",
                title = "Predicting the ultra-low frequency plasma wave power using solar 
                         wind data: a neural network approach",
                 year = "2022",
         organization = "COSPAR Scientific Assembly, 44.",
             abstract = "Changes in the configuration of the suns magnetic field influence 
                         the properties of the solar wind and, consequently, all planets 
                         and spacecraft within the heliosphere. Amongst other effects, 
                         perturbations in the solar wind generate waves within the Earths 
                         magnetosphere that can interact with energetic particles trapped 
                         within the Earths magnetic field. Ultra-low frequency (ULF) waves 
                         in Earths magnetosphere transport and energize energetic electrons 
                         in the Van Allen outer radiation belt via radial diffusion. The 
                         main goal of this work is to conduct a statistical study of ULF 
                         wave occurrence patterns using ground-based magnetometer data at 
                         high latitudes and thereby estimate the power spectrum density of 
                         these ULF waves, which is needed to model the radiation belts. We 
                         also use observations from the solar wind at the L1 Lagrangian 
                         point over the course of two solar cycle phases. Finally, we use 
                         Recurrent Neural Networks to predict the ULF integrated power at 
                         latitudes that can be mapped to the Van Allen outer radiation 
                         belt. Therefore, this work helps improve estimates of the 
                         radiation belt electron diffusion coefficients corresponding to 
                         ULF waves, a crucial factor in any particle diffusion models for 
                         the outer radiation belt.",
  conference-location = "Athens, Greece",
      conference-year = "16-24 July 2022",
             language = "en",
           targetfile = "PRBEM.3-0006-22-nopref.pdf",
        urlaccessdate = "30 abr. 2024"
}