AIRBORNE CAMPAIGN OBSERVATIONS AND ANALYSIS FOR THE CONFIGURABLE SCANNING SUBMILLIMETER-WAVE INSTRUMENT/RADIOMETER (COSSIR)
Marzo 25, 2026MACHINE LEARNING ENHANCED SNOWFALL RATE RETRIEVALS FROM PASSIVE MICROWAVE SENSORS
Marzo 25, 2026P. McEvoy1, E. May1, P. Eriksson1
1Chalmers University of Technology
The recently launched Arctic Weather Satellite (AWS) carries a radiometer that measures atmospheric temperature and humidity through conventional 54, 83, 183 GHz frequency bands, while also featuring novel 325 GHz channels. This makes AWS the first operational weather satellite to measure in the sub-millimetre band, enabling new retrieval and detection capabilities for clouds, ice, and snow.
The sub-millimetre band has historically represented an unfilled gap between microwave and infrared frequencies in meteorological space-borne observation systems, leaving a range of atmospheric particle sizes unobservable. AWS and upcoming missions are filling this gap for passive observations, enabling new retrievals and other applications for use in weather forecasting and climate modelling.
To properly utilize these measurements, the underlying physical processes must be understood and modelled. We present our work on atmospheric radiative transfer simulations utilising our database of microwave and sub-millimetre scattering properties for hydrometers and compare this with real AWS data. Our simulations demonstrate excellent statistical agreement with the real observations, though disagreement over ice and snow covered land motivates further development of surface-emissivity modelling for these cases. Comparisons between simulations and observations also reveal some cases of missing microphysical assumptions, valuable insights made possible by the sub-millimetre channels from this new instrument.
Despite these challenges, simulations show very good agreement for the vast majority of atmospheric cases. We demonstrate two applications made possible by this new frequency band and corresponding simulations: a more accurate cloud-masking technique that can more robustly and precisely distinguish humidity from clouds, and physically-based retrieval of ice masses in the atmosphere, which constitute an Essential Climate Variable as designated by the World Meteorological Organization.
Looking ahead, insights gained from this work can be applied to future missions. Most immediately, the proposed EPS Sterna constellation will consist of six satellites based on the AWS specification operating in polar orbit for 13 years. Additionally, the upcoming Metop-SG-B2, carrying the Ice Cloud Imager radiometer sensitive to frequencies from 325 to 664 GHz which enables more capabilities such as vertically resolved ice masses, will be operating for 21 years. Together, these missions promise a wealth of future sub-millimetre data to exploit.
