Authors: A. Le Pichon¹, M. De Carlo¹, L. Ceranna², E. Marchetti³, A. Hauchecorne⁴, S. Khaykin⁴, P. Espy⁵, R. Hibbins⁵

1. CEA/DAM/DIF, F-91297, Arpajon, France
2. BGR, Hannover, Germany
3. Departement of Earth Sciences, University of Firenze, Florence, Italy
4. LATMOS/IPSL, Guyancourt, France
5. Norwegian University of Science and Technology, Department of Science, Norway

The characterization of coherent infrasound as recorded by the International Monitoring System (IMS) infrasound network is important for assessing the detection probability of specific signals of interest. We systematically characterize coherent infrasound recorded by the IMS network over 10 years on 41 stations over a broad frequency range (0.01-5 Hz). This multiyear processing emphasizes continuous signals such as mountain associated waves and microbaroms, as well as persistent transient signals such as repetitive volcanic signals, thunder, or anthropogenic activity. For most IMS arrays, the detection of persistent sources is controlled by the dynamics of the stratospheric wind circulation from daily to seasonal scales. Analyzing long-term detections from well identified sources provides additional constraints on the dynamics of the middle atmosphere where data coverage is sparse. In order to account for a realistic description of the dynamic structure of the atmosphere, model predictions are further enhanced by wind and temperature error distributions as measured by high-resolution middle atmospheric sounding techniques.