Shu-Chih Yang
Global Atmospheric Observation and Data Applications Research Center, NCU
Bias characteristics and estimation of the FORMOSAT-7/COSMIC-2 radio occultation bending angle in the deep troposphere with a machine learning algorithm
FORMOSAT-7/COSMIC-2 radio occultation (RO) measurements have great potential in monitoring the deep troposphere and offering crucial insights into the Earth’s planetary boundary layer. However, the RO data retrieved from the deep troposphere can have severe bias under specific thermodynamic conditions. This bias originates from the limitations of the retrieval technique, the assumptions used in the algorithm and atmospheric influences. This study examines the characteristics of the RO bending angle bias (BAB). Based on those characteristics, this study proposes a machine learning-based algorithm, which uses a multi-layer perceptron neural network trained with different input proxies to assess region-dependent BAB. The results show that the BAB model is adequate to accurately predict the BAB in the deep troposphere in different regions. This research highlights the promise of advanced methodologies in improving RO retrieval and promotes data applications in the lower atmosphere.
