Introductione Advanced Model for Atmospheric TeraHertz Analysis and Simulation (AMATERASU) [1] is a Python program for analyzing THz systems used for atmospheric remote-sensing or wireless communication. Given atmo-spheric conditions, it simulates how the radiation propa-gates through the atmosphere and the level of attenuation. It has been used to various types of research. For instance, it helped to verify a model proposed for correcting the troposphere induced delays on the 2 and 14 GHz com-munication channels used for the ESA’s Atomic Clock Experiment in Space (ACES) [2]. It has also been used to analyze atmospheric observations near 0.6 THz with space-borne instruments [3]-[5], and to process the CO2 mea-surements of the coherent infrared Lidar (2 µm) built in the National Institute of Information and Communications Technology (NICT) [6]. It was the core of the NICT pro-cessing chain for the Superconducting Submillimeter-Wave Limb-Emission Sounder (SMILES) which operated during the winter 2009-2010 from the Japanese Experiment Module (JEM) onboard the International Space Station [5]. e mission was jointly led by the Japan Aerospace Exploration Agency (JAXA) and NICT with scientic PI from Kyoto University. It measured the atmospheric radia-tion spectra near the ozone (O3) line at 625 GHz. Using the small Doppler-shi signal in this line, we have retrieved winds in the altitude range between 35 and 80 km where no global observations are available [7]. Such data are ex-pected to improve meteorological and climate models as, 1AMATERASU（テラヘルツ大気放射伝達モデル）は、大気リモートセンシングや通信のテラヘルツシステムの設計や解析に利用するために開発された。現在、中層・上層大気の観測を目的とする衛星ミッション、スウェーデンのSIWと日本のSMILES-2の検討に用いられている。この2つの衛星ミッションはどちらも、中層大気の風を宇宙から初めて観測することを目的としている。この検討のために、AMATERASUの新たなモジュールを開発した。そのモジュールでは、オープンソースのTensorFlowライブラリを使うことでGPUによる計算を高速化し、さらに、地磁気によるゼーマン効果によって生じるテラヘルツ波大気伝搬の偏波依存性を解析することを可能とした。この論文では、開発内容とともに、現在進行中の衛星ミッションについてのこのモジュールに関連した検討状況を報告する。The Advanced Model for Atmospheric TeraHertz Analysis and Simulation (AMATERASU) has been developed for designing and analyzing THz systems for atmospheric remote sensing or wireless communication. It is currently used to study two satellite missions for the middle and up-per-atmosphere: the Swedish SIW and the Japanese SMILES-2. They both aim at measuring, among other, middle-atmospheric wind for the first time from space. To carry out these studies, new AMATERASU modules have been implemented. They allow us to accelerate the computation on GPUs using the open source library TensorFlow, and to consider the radiation polarization and the Zeeman effect induced by the geomagnetic field. In this paper, we describe these new devel-opments within the context of the on-going satellite mission studies. 4-5 AMATERASU（テラヘルツ大気放射伝達モデル）の新機能 : SIWとSMILES-2ミッションのためのGPUによる多偏波放射伝達計算の高速化4-5New Features in AMATERASU: GPU Accelerated Polarized Radiative Transfer for SIW and SMILES-2 Studiesバロン フィリップPhilippe BARON1074 衛星センサによる宇宙からの地球環境観測
元のページ ../index.html#111