Investigating the Use of Machine Learning (ML) to Assess Tropospheric Doppler Radar Wind Profiler (TDRWP) Data Quality

The Terrestrial and Planetary Environments (TPE) Team within the Natural Environments Branch (NEB) at Marshall Space Flight Center (MSFC) utilizes TDRWP data to provide input to space launch vehicle loads and trajectory calculations, ensuring that appropriate quality control (QC) measures are implemented before providing the data to end users. Previously documented QC processes entail executing automated and manual checks, and post-processing the resultant dataset for end-user needs. The manual QC checks, while allowing one to identify suspect or erroneous data subjectively, has been found to be time prohibitive. Thus, the use of ML is being explored with the intent to automate the attributes of the manual QC process. A summary of a trained ML model and its performance to classify erroneous data will be summarized along with future efforts to improve it.