TY - RPRT
AU - Katie Lean
AU - Niels Bormann
AB - The advent of space-borne hyperspectral microwave (MW) instruments brings a major change to MW sensor technology that has been relatively stable for many years. The present project develops for the first time the use of hyperspectral MW data in a global Numerical Weather Prediction (NWP) system, and the expected impact will be assessed using the Ensemble of Data Assimilations (EDA) method. This report considers necessary assimilation choices and outlines characteristics of the instrument to be used. One-dimensional information content studies have previously shown that a hyperspectral MW instrument has the potential to exceed the benefit in clear and cloudy temperature and humidity retrievals relative to its traditional counterpart. The wide frequency band and fine spectral sampling result in a diverse vertical sensitivity and exploitation of novel parts of the MW spectrum including the ability to probe narrow features in the oxygen absorption band around 50-60 GHz. To manage the large data volumes in assimilation, forming a channel subset will be required and an approach guided by the current strategy for hyperspectral infrared instruments is proposed. After initial removal of channels e.g. through quality control, channel selection following an information content-based approach may be employed. The Hyperspectral MW Sounder (HyMS) developed by Spire Global Inc. (Spire) and Rutherford Appleton Laboratory Space (RAL Space) will form the basis of the instrument to be evaluated. This covers a temperature (49.42-57.75 GHz) and humidity band (182.31–190.31 GHz) in addition to a window channel at 89 GHz. The fine spectral sampling means that individual channels have relatively high instrument noise, which needs to be compensated for by using a larger number of channels and, in some limited cases, averaging neighbouring frequencies to form a low-noise super channel. Adaptations in the all-sky MW assimilation strategy for HyMS are considered which include restriction of surface-sensitive channels over land/sea-ice and frequency selection for quality control procedures. The all-sky observation error model also requires modification to choose suitable indicators to estimate the presence of cloud and account for inter-channel error correlations.
BT - ESA Contract Report
CY - Reading
DA - 05/2025
DO - 10.21957/2752effd7a
M1 - 4000145264/24/NL/IB/ab
M3 - ESA Contract Report
N2 - The advent of space-borne hyperspectral microwave (MW) instruments brings a major change to MW sensor technology that has been relatively stable for many years. The present project develops for the first time the use of hyperspectral MW data in a global Numerical Weather Prediction (NWP) system, and the expected impact will be assessed using the Ensemble of Data Assimilations (EDA) method. This report considers necessary assimilation choices and outlines characteristics of the instrument to be used. One-dimensional information content studies have previously shown that a hyperspectral MW instrument has the potential to exceed the benefit in clear and cloudy temperature and humidity retrievals relative to its traditional counterpart. The wide frequency band and fine spectral sampling result in a diverse vertical sensitivity and exploitation of novel parts of the MW spectrum including the ability to probe narrow features in the oxygen absorption band around 50-60 GHz. To manage the large data volumes in assimilation, forming a channel subset will be required and an approach guided by the current strategy for hyperspectral infrared instruments is proposed. After initial removal of channels e.g. through quality control, channel selection following an information content-based approach may be employed. The Hyperspectral MW Sounder (HyMS) developed by Spire Global Inc. (Spire) and Rutherford Appleton Laboratory Space (RAL Space) will form the basis of the instrument to be evaluated. This covers a temperature (49.42-57.75 GHz) and humidity band (182.31–190.31 GHz) in addition to a window channel at 89 GHz. The fine spectral sampling means that individual channels have relatively high instrument noise, which needs to be compensated for by using a larger number of channels and, in some limited cases, averaging neighbouring frequencies to form a low-noise super channel. Adaptations in the all-sky MW assimilation strategy for HyMS are considered which include restriction of surface-sensitive channels over land/sea-ice and frequency selection for quality control procedures. The all-sky observation error model also requires modification to choose suitable indicators to estimate the presence of cloud and account for inter-channel error correlations.
PB - ECMWF
PP - Reading
PY - 2025
T2 - ESA Contract Report
TI - Evaluation of Hyperspectral MW for NWP: Simulation framework consolidation
UR -  
SN - 4000145264/24/NL/IB/ab
ER -